Jekyll2023-09-22T01:58:41+00:00https://wolframalexa.github.io/feed.xmlAlexa JakobThe <a href="http://hyde.getpoole.com" target="_blank">Hyde</a> theme for <a href="http://jekyllrb.com" target="_blank">Jekyll</a>, refreshed.Alexa JakobGoodbye Formlabs, Hello LeafLabs2023-09-21T00:00:00+00:002023-09-21T00:00:00+00:00https://wolframalexa.github.io/blog/2023/09/21/new-job-leaflabs<p>I recently started a new job at LeafLabs, an engineering services company, where I do a lot of PCB/electrical design. I’m really excited about it - the team is talented, smart, kind, and experienced, and I’m excited to work with them.</p>
<p>I spent around a year at Formlabs, working on the next-generation SLA printer - and I’m excited to share more when that product is released. I really enjoyed being part of Formlabs’s electrical engineering team, and learned a lot from my colleagues both inside and outside of engineering. It’s been a lot o’ learning, and I’ve written up some of the key takeaways!</p>
<p>Some of the main things I learned at Formlabs:</p>
<h2 id="building-scalable-products">Building scalable products</h2>
<p>Designing something for tens of thousands of units is very different than designing 10 units or turning in homework for something that will never be manufactured. When products are being made by contract manufacturers, there is an expectation of repeatability and reliability in design. Going through the new product introduction (NPI) process taught me about designing for scale, communicating with overseas vendors and manufacturers, and anticipating common pitfalls in the manufacturing process and trying to fix those early on with design and documentation.</p>
<h2 id="managing-a-very-technically-complex-project-and-hitting-tight-deadlines">Managing a very technically complex project and hitting tight deadlines</h2>
<p>This goes hand in hand with ruthless prioritization - what needs to be investigated now, vs what can be left to later? What do we want to supervise in the field, vs what should be tested in-house? What can we lock in now for the design, vs what do we want to leave flexible and decide further down the line? Where can we spend money to make a problem go away, vs what is worth the time to cost-down? My main design area required collaborating with nearly every engineering discipline at Formlabs, so I got to know a lot of people across the organization. Mechanical and electrical engineers tend to think very differently based on our tools and processes (maybe this is a future blog post), so I had to understand their point of view to weigh different goals and timelines to deliver an integrated final product.</p>
<h2 id="advocating-for-my-ideas-and-getting-resources-for-what-i-need">Advocating for my ideas and getting resources for what I need</h2>
<p>It’s very easy to think that everything is important all the time, but realistically there are only 40 hours in the work week and only one of me, so prioritization is necessary. In the same vein, developing hardware takes far more than just doing the schematic/layout capture - there’s a lot of meetings to gather requirements, testing, writing documentation, answering questions, and debugging problems when your hardware is out in the world. I found that being honest with my colleagues about capacity and pointing to key decisionmakers for prioritization to be very helpful in managing my time as an individual contributor. Similarly, in advocating for my point of view in technical discussions, tying your preferred path to a stated goal (cost-down, reliability, performance, etc) and qualifying its impacts on existing products is useful.</p>
<h2 id="having-compassionate-leadership-is-irreplaceable">Having compassionate leadership is irreplaceable</h2>
<p>We all learn from making mistakes. Early on, during verification of new hardware, I had made a mistake that damaged some boards and apologized profusely to my manager, who said “well, are the boards blowing up by themselves, or are you doing that?” I sheepishly admitted it was my fault, and he told me that as long as it wasn’t a systemic design issue, it was fine, and encouraged me to be more careful and take my time. Over the course of my time at Formlabs, I also appreciate how Jon gave me very useful feedback, advocated for me to other leaders in the organization, and proactively looked for opportunities for me.</p>
<p>Overall, this was my first engineering job out of college, and my team put a lot of trust in me and gave me challenging and serious work from day one. When I told a colleague I was leaving, he told me “you’ve definitely made your mark on this program.” That was a goal of mine, and I’m proud to have accomplished it. I’ll always have a lot of appreciation for my Formlabs colleagues and the friends and mentors I met there, and I’m excited to see what the company does next and share more about my role in it when the new product is released!</p>
<h1 id="that-said">That said,</h1>
<p>there’s a lot to be excited about at LeafLabs!</p>
<h2 id="working-on-a-wide-variety-of-projects-and-solve-really-hard-client-problems">Working on a wide variety of projects and solve really hard client problems</h2>
<p>LeafLabs’s clients really run the gamut of organizations - from small startups to Fortune 500 companies - and I’m excited to have a glimpse into many different hardware stacks, processes, and cultures. Companies bring in LeafLabs when they need our expertise, which means we get to solve a lot of difficult and critical challenges, and I take the responsibility of being trusted with these challenges seriously. It’ll be fun and rewarding! Also, engineering ethics matters very much to me, and LeafLabs never works on destructive military or financial applications. It feels really good to have my values reflected in the organization I work for.</p>
<h2 id="deepen-my-skills-across-the-electrical-and-embedded-engineering-stack-by-learning-from-colleagues">Deepen my skills across the electrical and embedded engineering stack by learning from colleagues</h2>
<p>Many of my colleagues have expertise at multiple organizations in embedded firmware, FPGA development, hardware design, and more. They’re cross-functional computer engineers who know a <strong>lot</strong> about how embedded systems are built and used. Embedded and firmware is an area I want to grow in, so I’m really excited to learn from them. At the same time, I’m an electrical engineer by training, so I’m excited to bring that perspective to the organization and work on developing new and scalable hardware products.</p>
<h2 id="working-on-more-research-oriented-problems">Working on more research-oriented problems</h2>
<p>Neuroscience research is a big focus of LeafLabs, our flagship project being <a href="https://www.leaflabs.com/willow">Willow</a>, a 1000-channel data acquisition system for data acquisition. I’m excited to be working the next generation of neuroscience at LeafLabs, making more data more available and easier to gather for neuroscience researchers. This is a space that’s ripe for innovation and one that the NIH is particularly interested in, so it feels very rewarding to be working on technology for the public interest.</p>
<p>I’m only a few weeks in, but the work has already been rewarding and my new colleagues are really funny and helpful and nice. I’m excited for what’s to come!</p>Alexa JakobSwitching out the 'labs' prefixDon’t Rely on Diversity in Tech to Fix AI Ethics2023-03-05T00:00:00+00:002023-03-05T00:00:00+00:00https://wolframalexa.github.io/blog/2023/03/05/diversity-in-ai-not-silver-bullet<p>I went to <a href="https://nightofideas.org/boston/">Night of Ideas</a>, a yearly series of philosophy lectures held in 20 cities across the United States. This year’s theme was “More?”, and I attended the lecture titled “ChatGPT, the Metaverse, and AI: How will they be affecting education?”. It was insightful - as a general rule, I’m skeptical of hype surrounding new technologies like ChatGPT, and I want us to be thoughtful about how they’re implemented - but Prof Katz detailed some realistic ways that ChatGPT could ease the burden of educators, for example by drafting response to parent emails and creating associated follow-up tasks.</p>
<p>Both sections (ChatGPT and the metaverse) addressed the pressing issue of bias in AI, but the solution presented was a more diverse tech workforce, the idea being that more women working on ChatGPT will make it less sexist, and so on. I find this idea - that more diversity in tech will be a silver bullet for non- or anti-biased AI - very common in tech spaces, and I’m skeptical.</p>
<h2 id="this-will-not-probably-work-because-of-how-the-technology-is-structured">this will not probably work because of how the technology is structured</h2>
<p>ChatGPT and other large language models are trained on data from the internet, and surprise, there is a lot of bias on the World Wide Web. Could you change your training data? Sure, but you need <em>so much data</em> that the internet is really the best option - so how much of a change could your engineers actually make? (That is also a different solution than hiring diverse engineers.) ChatGPT for example, has some safeguards built in, but <a href="https://www.bloomberg.com/news/newsletters/2022-12-08/chatgpt-open-ai-s-chatbot-is-spitting-out-biased-sexist-results">people have found ways to override even those</a>.</p>
<h2 id="you-will-certainly-miss-some-things">you will certainly miss some things</h2>
<p>You can hire a diverse team and still miss spots. There are many axes of harm: gender, sex, race, age, ability, national origin, and more. There is no conceivable way a team of 10-15 people could encompass the entire human experience. If you’re relying on your engineers to bring up potential points of bias, you will certainly miss some. Those ‘diverse’ team members don’t necessarily understand the harm that your systems are doing to people of their marginalized group simply by virtue of their life experiences (believe it or not, not all women, or Black people, or disabled people, etc. are the same!), and further, you might be asking these people to relive traumatic experiences. There is no way you can catch every single harm, and you might be causing more harm in the process.</p>
<h2 id="youre-harming-your-diverse-team-members">you’re harming your “diverse” team members</h2>
<p>One of my tenets in the “women in engineering” space is that women engineers should be able to be respected and valued as engineers while being held to the same performance standards as their male counterparts. This seems obvious and like everyone can agree, and should be true for every marginalized group. That said, by relying on the BIPOC employees on your team to build anti-racism into a system, you are <strong>fundamentally asking them to do a different job, with different responsibilities, than their white counterparts</strong>. How this usually shakes out is that marginalized employees do a lot of the “<a href="https://noidea.dog/glue">glue work</a>” of keeping the team functioning - training their coworkers on bias, mentoring others, setting up systems - while privileged employees do the highly visible and promotable work of coding. This too is labour - but it’s not part of their job description.</p>
<p>In hiring someone so they can “debias” your product, you’re essentially saying that their demographic characteristics matter more than their technical skills. You’re applying a different standard to underrepresented people just because of their identity, which does not respect them neither as people, nor as engineers. Making marginalized people solely responsible for preventing harm is not fair, and it’s the wrong way to build a diverse team that respects all viewpoints and identities.</p>
<h2 id="whats-better-a-systematic-approach">what’s better? a systematic approach</h2>
<p>Ultimately, I want to see tech companies move towards a systematic approach to ethics and addressing bias, rather than asking individual teams to identify issues ad hoc. Hiring ethics experts whose job it is to review design proposals, consult on the design process, and develop a rigorous testing process before anything is released (aka putting your money where your mouth is), is far more fair and will result in catching more potential harm before it happens. There should be precise goals and check-in points that allow for feedback and correction, early and often.</p>
<p>To be sure, there have been some criticisms of the way these teams have been implemented (and <a href="https://www.engadget.com/meta-responsible-innovation-team-disbanded-194852979.html">they don’t always last long</a>) — but I still think having people engaged on these issues, rather than shoehorning it onto the plate of the few minority engineers, is preferable.</p>
<p>Avoiding bias is hard. Designing systems to be non-biased takes a lot of education, thought, and commitment - it’s labour. Let’s pay the experts.</p>Alexa JakobAsking marginalized tech workers to save us is not a good strategySome Thoughts on Hanukkah of Data 2022/57832023-02-18T00:00:00+00:002023-02-18T00:00:00+00:00https://wolframalexa.github.io/blog/2023/02/18/hanukkah-of-data<p>I heard about <a href="https://hanukkah.bluebird.sh/5783/">Hanukkah of Data</a> on Twitter, and thought it looked fun. I’m trying to get better at using data tools in Python since analyzing data is a big part of my job, as it is for any engineer - qualifying parts, testing design changes, looking at how customers currently use our products so that we can make better decisions in design, that sort of thing. My code is <a href="https://github.com/wolframalexa/HanukkahOfData">here</a> and of course contains spoilers!</p>
<p>In Hanukkah of Data, you’re given several databases relating to customers, orders, and items at Noah’s Deli, a fictional deli in New York City. In every case, you use your detective/data skills to identify a past customer who was once in possession of a rug in order to track it down.</p>
<p>I found the challenge to be pretty approachable to beginners. There are a lot of memes about how programming is basically just stringing together the correct StackOverflow answers until you get something that works, and this was sort of true here. The questions were structured in a way that I could easily identify several characteristics to look for (an Aries, say, or someone who lived in Ozone Park), so I could look up “how to XYZ in pandas”. If I had spent more time, or had more experience, I probably could have written more sophisticated code that joined multiple tables in one line, or imported the databases into SQLite to query them using SQL, but my code is readable, which feels like the real win here.</p>
<p>I also appreciated that the bulk of the puzzle was in the conceptual understanding of the question. In <a href="https://adventofcode.com/">everyone’s favourite other holiday-themed coding challenge</a>, the answer’s implementation often feels like a drag after I’ve conceptually solved the puzzle. Here, the implementation felt trivial, and the code was far more forgiving since errors were more easily spotted (who amongst us has not spent hours debugging an off-by-one error to enable the elves to use warp drive or something).</p>
<p>One thing I’d like to see more of would be more tricks. For example, in puzzle 8, the person who owned all the collectibles was also the person who had the most collectibles, which was far easier to filter for than “owned all the collectibles”. It would be very strange if someone owned multiple of the same collectible, but it certainly would be possible (who knows, maybe Noah’s has some collectible resale market!). Introducing elements that are unexpected could be a fun way to stretch one’s data skills further.</p>
<p>That said, for the first year running this challenge, I really enjoyed it and found it to be really enjoyable and approachable. I was away from my laptop over the holidays, and I was able to do most of it in one day when I returned. I know people who spend hours on AoC every day at midnight, and that’s just not the life I’m trying to live. I’m excited for more cute bagel-themed puzzles next year!</p>Alexa JakobWho doesn't love detective work and bagels?A Very Belated 2022 Recap2023-02-05T00:00:00+00:002023-02-05T00:00:00+00:00https://wolframalexa.github.io/blog/2023/02/05/end-of-year<p>This is a very abridged version of the end-of-year message I sent out to my friends and colleagues. This email is somewhat of an experiment for me - I’m inspired by my friend Tommy, who every year writes an email reflecting on his year, and my parents, who every year print and mail a letter with my and my siblings’ pictures and accomplishments in it. New Year’s brings a lot of thoughts and reflections about the year that we’ve had. This post of course is incredibly belated. But there’s nothing inherently special about January 1st other than the vibes we assign to it so I think a February retrospective is just as nice.</p>
<p>I’ve had quite a busy 2022! Here’s the highlight reel:</p>
<h1 id="competed-on-jeopardy">Competed on Jeopardy!</h1>
<p>This was easily one of the coolest experiences of my life! I got to meet Ken Jennings, visit the Jeopardy! and Wheel of Fortune sets (the wheel is way smaller in real life), and made some wonderful friends out of the whole experience.</p>
<p>It was interesting being there as one of the younger contestants on regular Jeopardy!. My tape-date-mates were doctors, lawyers, writers, professors, and a court reporter, and I taped my episode not three days after having finished my last final of college (and with a mountain of grading to do for my computer architecture TA gig). It was certainly intimidating, and I don’t remember much of the actual taping, though I’m told I came across as confident and cool. When I watched the show, I wondered who would buzz in on the next question - surprise, it was me, and I got it correct! I did lose, but <a href="https://www.youtube.com/watch?v=BvUZijEuNDQ">there’s a great Weird Al song about that</a>.</p>
<p>When the show aired, the Internet was, well, the Internet. I still think it was a net positive, though - I gained a lot of confidence in myself and in public speaking through the experience. The real win was the friends we made along the way, though - the Jeopardy! alumni community is strong, and it’s helped me appreciate the power of social media to bring people together. I have met up with a few J! folks in Boston and hope to do more of that soon. I’m still really bad at pub trivia, though.</p>
<h1 id="biked-to-albany-for-climate-action">Biked to Albany for climate action</h1>
<p>Sunrise Movement NYC, a climate organizing group I have been active with for the past two years, joined the Climate Can’t Wait NY coalition to push for climate legislation in New York State. Veekas, one of our organizers, conceived of a bike trek from NYC to Albany along the Empire State Trail leading up to an Earth Day rally in Albany. I was initially skeptical, but decided to take the week off school and do it anyways. I’m very glad I did — I’m proud of the work we did to raise the profile of this legislation, and we were able to connect with people across the state who believe in climate action and are willing to work to make it happen. I have never felt very comfortable as a “climate leader” — I don’t think I know enough about the science or policy solutions to be put front and center — and this was a key step in realizing you don’t have to know it all to speak up, because the more people who speak up in favour of climate action, the better a shot we have at literally saving life on planet Earth.</p>
<p>This trip was also meaningful to me as a cyclist. This was the farthest I’ve ever biked after only starting biking the previous August, and I got a lot stronger as a result. I’m starting to see myself as an athlete, a big shift for someone who has always hated the gym. The trail was beautiful, but I did unfortunately learn the importance of eating correctly and drinking enough water the hard way. Still, that trip gave me the confidence to bike longer distances and more consistently, and I completed my first century (100 miles) in June.</p>
<h1 id="graduated-college">Graduated college</h1>
<p>I now have a bachelor’s of electrical engineering from the Cooper Union with minors in computer science and philosophy! My college experience was far from typical, due to the pandemic, yes (COVID hit halfway through my sophomore spring and caused just over 1/4 of my classes to be delivered remotely), but also due to how Cooper Union is structured as an institution. The deep investment of faculty in students, the history of free tuition, and the simultaneous specialization and breadth of its offerings make attending Cooper an unusual college experience. I mean that in the best way possible - I benefitted from the strangeness, and I don’t think I would have felt nearly as comfortable expressing myself, or challenged to push research questions further, at a more traditional institution. At times, it was isolating and discouraging too. I am so thankful to the faculty and staff who created space for me to explore new ideas, promoted my work to their colleagues, and created opportunities for me. They went beyond mentorship and became <a href="https://fairygodboss.com/career-topics/mentors-vs-sponsors">sponsors</a>, and they have changed my academic and career trajectory for the better.</p>
<h1 id="started-a-new-job">Started a new job</h1>
<p>Speaking of career prospects, I started a new job in August. I’m working as an electrical engineer at Formlabs, a 3D printing company based in Somerville, MA. It’s interesting work - I get to design circuits for our next generation SLA printer, and I work with mechanical, software, optical, and integration engineers to get that done. Although working full-time has been a big shift compared to college, I really appreciate the steady paycheck and going home at 5PM!</p>
<p>My manager has been great about putting me on new and interesting projects, and I am learning so much. My colleagues take me seriously and value my contributions, even though I am new. One anecdote about the kind of work culture it is - when we design circuit boards, we put our initials on them (so the ones I design say “AJ”). One day, I helped out our EE lead, Sam, with some layout on a circuit board, and didn’t think much of it. When he went to place the order for the boards, I saw that he had added my initials next to his. I told him that I appreciated it and thought it was kind, and he said that of course I was credited - I designed the board.</p>
<h1 id="moved-to-boston-and-met-a-lot-of-new-people">Moved to Boston and met a lot of new people</h1>
<p>I moved for work (see above). Moving is always hard, but I feel like I’ve mostly landed on my feet. I see friends, new and old, fairly often and I enjoy my work. I have found little communities in different places, like social dance and swimming. Boston feels less like the centre of the universe than NYC does, but I don’t need that to be content —I like being around interesting people and places, and it’s been rewarding finding local niches here. I didn’t spend much time going to the Met anyways.</p>
<h1 id="lots-of-personal-growth">Lots of Personal Growth</h1>
<p>It was a good year for me learning to take care of myself. I started sleeping better and more consistently, eating more healthfully, and getting outside more. Quantitative measurements don’t do it justice, but I biked 2185 miles and read 40 books. My interpersonal relationships got stronger, and I became better at making friends.</p>
<p>Though I experienced a lot of change, it was the result of active choices I made to align my life more with long-term goals, or in some cases, to follow what just felt right in the moment. And the cool thing is, making difficult choices (i.e. uprooting my life and moving to a different city) has proven to me that big changes are not inherently scary or harmful, and I believe I would be more comfortable making a similarly large change again if it would improve my life, rather than feeling stuck in the way things are. It’s a virtuous cycle, and I’m excited to see where 2023 leads me.</p>
<p>There’s so much I couldn’t fit in this post that made an impact on my 2022, but it was a year well-lived. I’m grateful for all the people, organizations, and things that made it special. Here’s to 2023!</p>Alexa JakobIn which I try to make sense of the madness that was 2022To Alleviate the Electrical Engineer Shortage, Invest in Diversity and Inclusion2022-07-27T00:00:00+00:002022-07-27T00:00:00+00:00https://wolframalexa.github.io/blog/2022/07/27/ee-shortage-diversity<p>There’s been an interesting discussion in the tech world lately about the relative unpopularity of electrical engineering compared to computer science, after <a href="https://www.theregister.com/2022/07/18/electrical_engineers_extinction/">these</a> <a href="https://www.theregister.com/2022/07/08/semiconductor_engineer_shortage/">articles</a> were published in the Register.</p>
<p>Essentially, the proportion of electrical engineering graduates has sharply declined in recent years, while the proportion of computer science graduates has increased. Even this decline in graduates doesn’t tell the whole story, since many electrical engineering graduates are choosing to work in software.</p>
<p>This is generally a problem as computers become bigger and more complex - the article notes that companies like Samsung, Intel, NVIDIA, and Qualcomm are making big investments in fabs that will need to be staffed by thousands of electrical engineers. Where are all the electrical engineers?</p>
<p>I’m inclined to agree with many of the points that Goodwins makes. He notes that there is less exposure to electronics now than when he was growing up, so kids don’t catch the “tinkering bug” like they once did. After all, Radio Shack is no more, and a transistor radio is more approachable than an iPhone.</p>
<p>Software jobs are by and large more lucrative and possible to be done remotely, which makes them attractive to young engineers. In my own EE degree class, I estimate half of us became software engineers or data scientists - the rest pursued math, circuit design, radio frequency engineering, etc.<sup id="fnref:1" role="doc-noteref"><a href="#fn:1" class="footnote" rel="footnote">1</a></sup></p>
<p>While all these factors are important, diversity and inclusion are overlooked factors in the difference between EE and CS enrollment. I believe that as electrical engineers, embracing D&I will help our profession grow.</p>
<p>In past years, there has been a remarkable growth of organizations designed to funnel women and other underrepresented minorities into computer science. Organizations like Girls Who Code, Rewriting the Code, ColorStack, Black Girls Code, and Code2040 have created opportunities for people who might not otherwise see themselves in the technology industry. These organizations play a different role than legacy diversity in engineering organizations like the Society of Women Engineers, the National Society of Black Engineers, or the Society of Hispanic Professional Engineers, due to how they are structured.</p>
<p>Take Girls Who Code as an example. Their flagship program is a summer intensive, where girls in high school spend the summer learning how to code. The program is free (sponsored and hosted by a technology organization like Google, Cisco, or MetLife, who can also recruit from the program’s alumni), and each group of about 20 girls are taught web development for six weeks. The program aims to both teach skills and create community. The girls come away with a portfolio website, an understanding of common tech tools like GitHub and the command line, and the “sisterhood” of past, present, and future GWC students and teachers.</p>
<p>This is incredibly effective at introducing girls to computer science and encouraging them to pursue it in college. My own sister participated in a Girls Who Code program the summer before her junior year. She has just finished her first year of an electrical engineering degree and is teaching at the program this summer. She plans to pursue software engineering and partially credits GWC with her choice<sup id="fnref:2" role="doc-noteref"><a href="#fn:2" class="footnote" rel="footnote">2</a></sup>. And she’s far from alone - Girls Who Code has reached half a million girls and nonbinary youth in its history<sup id="fnref:3" role="doc-noteref"><a href="#fn:3" class="footnote" rel="footnote">3</a></sup>, and shows no signs of stopping now.</p>
<p>To my knowledge, a comparable program does not exist for electrical engineering. Many schools don’t teach engineering the way they do coding, and if you’re part of a minority group, being in a room with others like you could help you gain confidence. While SWE, for instance, does do outreach events at local schools or conferences, it does not have a sustained teaching program the way Girls Who Code does. Since the missions of SWE, NSBE, and SHPE are focused on serving all engineers of a certain minority group, it’s difficult for them to teach technical skills that will be applicable to many of their members. (The specific skills needed for civil engineering differ from those in robotics - and yet practitioners of these fields and many others all attend the same conference.)</p>
<p>Creating intentional communities that meet the needs of minority electrical engineers would be a great way to increase and sustain participation in electrical engineering. Compared with computer science, traditional engineering fields have far fewer women. According to Zippia, <a href="https://www.zippia.com/software-engineer-jobs/demographics/">22% of software engineers</a> are female, while only <a href="https://www.zippia.com/electrical-engineer-jobs/demographics/">10% of electrical engineers</a> are, among the lowest in engineering professions. To be sure, both of these numbers are low - but 10% is tiny!</p>
<p>It can be hard to feel like you belong when you don’t see others like you. In my (admittedly short) career, I’ve worked closely with only one woman engineer, and when I was looking for a full time job, I was not interviewed by any women in a technical interview. Obviously, it hasn’t stopped me, but I hope that someday we reach gender parity.</p>
<p>Anecdotally, attitudes in traditional engineering companies are more conservative than in tech. Since younger workers are seeing diversity and inclusion as more and more of a priority - you wouldn’t want to work where you’re not respected - creating an inclusive culture is essential. We know that discrimination drives talented women engineers to leave the field every year, at every stage<sup id="fnref:4" role="doc-noteref"><a href="#fn:4" class="footnote" rel="footnote">4</a></sup>, which is unacceptable when we have such a shortage.</p>
<p>Finally, but no less importantly, computer science seems to have a better public relations program. Tech billionaires, love them or hate them, talk often about the need to learn how to code, and associate coding with innovation and genius. Electrical engineering, while equally (or, I would argue, more) necessary, is less flashy.</p>
<p>Because big tech has money, they fund computer science departments at universities and programs like Girls Who Code to produce the next generation of their workers. For instance, my alma mater secured funding to start a new computer science major and build out a whole new program, with all the additional students and professors and funding that requires. The electrical engineering department was already catering towards students interested in software - those electives were overenrolled, while hardware electives were underenrolled and did not run. It doesn’t seem like the same kind of corporate funding exists to expand EE programs.</p>
<p>To alleviate the shortage of electrical engineers, we should recruit and retain engineers from diverse backgrounds. We could set up Girls Who Code-style programs to reach out to underrepresented youth and provide them with both skills and community. We could intentionally support engineers at every stage in their career journey. We could seek out funding to increase the number of EE seats at leading colleges and increase salaries to be on par with software roles.</p>
<p>I generally don’t like to make the business case for diversity, because I believe diversity should be a value of any organization, full stop - anti-discrimination is the law<sup id="fnref:5" role="doc-noteref"><a href="#fn:5" class="footnote" rel="footnote">5</a></sup>. That said, investing in diversity and inclusion can help alleviate the shortage of electrical engineers, in addition to it being the right thing to do. Besides salary and mathiness, let’s not overlook the need for inclusive environments as a driver of the electrical engineering shortage.</p>
<p><em>Many thanks to Jacob Kuppermann for proofreading this piece.</em></p>
<div class="footnotes" role="doc-endnotes">
<ol>
<li id="fn:1" role="doc-endnote">
<p>I’m part of the hardware half - I will be doing circuit & product design at my new job very soon! <a href="#fnref:1" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:2" role="doc-endnote">
<p>For what it’s worth, she says I deserve partial credit as well! <a href="#fnref:2" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:3" role="doc-endnote">
<p>https://girlswhocode.com/about-us <a href="#fnref:3" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:4" role="doc-endnote">
<p>This is called the leaky pipeline - and we need to start plugging all the holes. For more on why women leave engineering, see SWE’s research on the topic: https://alltogether.swe.org/2019/11/swe-research-update-women-in-engineering-by-the-numbers-nov-2019/ <a href="#fnref:4" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:5" role="doc-endnote">
<p>Susan Rigetti, the Uber whistleblower, introduced me to this concept. Essentially, creating a welcoming environment is not a “nice-to-have” - it’s required by the EEOC. You should not feel bad for exercising your right to a safe workplace <3 <a href="#fnref:5" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
</ol>
</div>Alexa JakobTry this one weird trick!Politics and Philosophy of the Crescent Street Bike Lane2022-06-13T00:00:00+00:002022-06-13T00:00:00+00:00https://wolframalexa.github.io/blog/2022/06/13/crescent-st-bike-lane<p>Last semester, I took a course called Philosophy of Infrastructure. As a major infrastructure and logistics nerd, I jumped at the chance, and it was a lot of fun! For our final project we had to “adopt a New York City infrastructure” and write about its ethical, epistemological, aesthetic, social, and political dimensions. I chose to write about a piece of infrastructure that I use nearly every day: the Crescent Street Bike Lane.</p>
<p>Biking down Crescent Street in Astoria, Queens, is easily my favorite part of my commute. It’s relatively quiet and protected from cars, and there are always children in the playground near the school it passes by. In many ways, the stretch of Crescent Street from the Queensboro Bridge to the Triboro Bridge embodies one of the most diverse neighborhoods in the world<sup id="fnref:1" role="doc-noteref"><a href="#fn:1" class="footnote" rel="footnote">1</a></sup> - not too far away, you can pass by the Mediterranean market on the way to one of the largest mosques in New York City. And yet, it is this diversity that makes the Crescent Street bike lane a site of conflict. In this essay, I argue that Crescent Street, particularly its bike lane, is the site of broader philosophical conflicts around safe streets, gentrification, and the broader New York City bike network.</p>
<h1 id="slides">Slides</h1>
<script async="" class="speakerdeck-embed" data-id="2b066c1665d048cdbe666d223d0c1f9b" data-ratio="1.77777777777778" src="//speakerdeck.com/assets/embed.js"></script>
<h1 id="bike-lanes-as-network">Bike Lanes as Network</h1>
<p>New York City is home to the largest bicycle network in the United States, comprising over 1,375 miles of bike routes <sup id="fnref:2" role="doc-noteref"><a href="#fn:2" class="footnote" rel="footnote">2</a></sup>. All of NYC’s bike lanes are administered by the New York City Department of Transportation (DOT), but it is far from a comprehensive network. The lanes do not go everywhere car lanes go, and they vary in the quality of protection they offer. When Kaika theorizes about the urban water and sewer system, she writes that “endowed with modernity’s technological networks, the urban fabric became a nexus of entry-exit points for a myriad of interconnected circuits and conduits.”<sup id="fnref:3" role="doc-noteref"><a href="#fn:3" class="footnote" rel="footnote">3</a></sup> Unlike other technological networks, the network of New York City bike lanes is porous – an impurity cannot enter the “good” water lines, but a car can too often enter a bike lane and cause harm. In unprotected lanes, the only thing keeping drivers from entering the bike lane is cultural norm – which is highly inconsistent because of the devaluation of the bicycle as a mode of transportation, and determined by individuals.</p>
<p>Due to these interruptions, both in the network itself and in the edges of the network, we might ask – what is safety, actually? How is this network responsible, or how does it absolve itself of responsibility for the safety of cyclists and motorists? How much agency can we assign to inanimate infrastructure? In the sense that bike lanes are designed by urban planners at the DOT, and sometimes brought about by community input, these artifacts certainly do have politics, and bike lanes are subject to both technological and social determinism. The “rules of the road” are always somewhat fungible, but traffic and bike iBnfrastructure help make them more explicit by controlling and formalizing existing behavior. In this way, safety is both a function of the physical barriers that separate cyclists from motorists, but also in the behavior and social relationships between the two.</p>
<h1 id="bike-lanes-as-safety-infrastructure">Bike Lanes as Safety Infrastructure</h1>
<p>In 2014, New York City committed to a policy of Vision Zero – a goal of achieving zero fatalities or injuries from traffic. In the years since, traffic deaths have actually increased, from 205 in 2018 to 273 in 2021<sup id="fnref:4" role="doc-noteref"><a href="#fn:4" class="footnote" rel="footnote">4</a></sup>. Protected bike lanes, like the one on Crescent Street, have been a key part of that policy.
The placement and grade of bike lanes is a political choice that revealing who deserves to travel safely in our cities, and who is expendable. In November 2020, a delivery worker named Alfredo Cabrera Liconia was fatally struck by a truck driver on Astoria Boulevard and Crescent Street<sup id="fnref:5" role="doc-noteref"><a href="#fn:5" class="footnote" rel="footnote">5</a></sup>. In the aftermath, politicians demanded upgrades, but they did not materialize until this year, when the Department of Transportation began “hardening” bike lanes (upgrading their level of protection), including Crescent Street.</p>
<p>As of the latest bike map<sup id="fnref:6" role="doc-noteref"><a href="#fn:6" class="footnote" rel="footnote">6</a></sup>, protected bike lanes, shown in green, are concentrated in Manhattan. Protected bike lanes mirror the subway map (Crescent Street itself runs parallel to the N/W trains in Astoria). This is a missed opportunity for transit justice – in areas where the subway does not run, like Eastern Queens, residents tend to drive to work or for errands. Providing a safe bike route, in places where public transit infrastructure is sparse, could help reduce car trips in those areas.</p>
<p>Equitable bike infrastructure is also a gendered issue. Women are more likely to carry the burden of household errands<sup id="fnref:7" role="doc-noteref"><a href="#fn:7" class="footnote" rel="footnote">7</a></sup>, whether bringing a sick child to the doctor or going to the grocery store. This travel occurs between neighborhoods, and is often badly serviced by existing “hub and spoke” transit infrastructure, in which riders are ferried to and from the urban core. Where women do cycle, they are more likely to be closely passed by cars and face more risk than male cyclists<sup id="fnref:8" role="doc-noteref"><a href="#fn:8" class="footnote" rel="footnote">8</a></sup>. Any safe streets policy must take these disparities into account.</p>
<h1 id="bike-lanes-as-locus-of-community-conflict">Bike Lanes as Locus of Community Conflict</h1>
<p>The idea for the Crescent Street bike lane came from community organizing. Advocates spoke to their neighbors, showed up to community board meetings, and got elected officials on board over a two-year long process<sup id="fnref:9" role="doc-noteref"><a href="#fn:9" class="footnote" rel="footnote">9</a></sup>. When the bike lane was finally built, advocates cheered, but not everyone was thrilled. In 2016, Astoria has had the dubious honor of being named the 11th most gentrifying neighborhoods in New York City,<sup id="fnref:10" role="doc-noteref"><a href="#fn:10" class="footnote" rel="footnote">10</a></sup> and 21.7% of households were severely rent burdened in 2019<sup id="fnref:11" role="doc-noteref"><a href="#fn:11" class="footnote" rel="footnote">11</a></sup>. These economic realities, coupled with changing demographics and more young white people moving in, create anxiety about gentrification and displacement among long-term residents.</p>
<p>In September 2021, Edwin DeJesus, a candidate for City Council with the Green Party, wrote an op-ed titled “I Love Biking Too, but the Crescent Street Bike Lane Is a Disaster.”<sup id="fnref:12" role="doc-noteref"><a href="#fn:12" class="footnote" rel="footnote">12</a></sup> In it, he argues against the existing bike lane, citing issues with parking, the privatization of bike share, and cars being forced to idle or block the bike lane. He cites the “interests of the community,” “our families,” and “life-long Astoria residents” to imply who he believes is to blame for these issues – “cyclists” and those who are new to the neighborhood.</p>
<p>DeJesus may be right to link bike lanes with gentrification. In “Forewarned: The Use of Neighborhood Early Warning Systems for Gentrification and Displacement”, Chapple and Zuk argue that bike lanes, among other transportation improvements, can be a warning sign of oncoming higher rents<sup id="fnref:13" role="doc-noteref"><a href="#fn:13" class="footnote" rel="footnote">13</a></sup>. It is true that such improvements might make a neighborhood more desirable, which can raise rents and push out existing residents – but does this mean that no safety improvements should be made? Must we accept higher cyclist and pedestrian deaths in disadvantaged communities? The bike lane protects the many delivery workers in Astoria who cycle for work, not only white newcomers, but the benefits aren’t perceived to be equally distributed. Besides, injuries decline for motorists, cyclists, and pedestrians when bike lanes are installed. Working-class neighborhoods deserve the safety that bike lanes provide, and these investments must be made equitably.</p>
<h1 id="conclusion">Conclusion</h1>
<p>Compared to what it once was, Crescent Street is more peaceful, due to fewer and slower cars. Longtime residents mingle with new arrivals, and the bike lane has helped make a street that was otherwise built to encourage speeding so safe that children ride bikes and scooters in it. Oftentimes, once the dust settles, investments in safe streets are appreciated by everyone. The ever-changing nature of the lane – from mere idea, to community board approval, to the most recent hardening phase – reflects the ways in which New York City’s infrastructure is always shifting and the philosophical questions that arise from NYC’s incomplete bike infrastructure, recent enough to not be “urban dowry,” but foundational and future-looking.</p>
<div class="footnotes" role="doc-endnotes">
<ol>
<li id="fn:1" role="doc-endnote">
<p>Alexandra Starr, “In New York’s Multinational Astoria, Diversity is Key to Harmony,” National Public Radio. March 30, 2015. https://www.npr.org/sections/codeswitch/2015/03/30/393339438/in-new-yorks-multinational-astoria-diversity-is-key-to-harmony <a href="#fnref:1" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:2" role="doc-endnote">
<p>“Biketober: Two-Way Protected Bike Lane Along Fort Hamilton Parkway Now Complete,” New York City Department of Transportation, October 22, 2021. https://www1.nyc.gov/html/dot/html/pr2021/biketober-two-way-protected-bike-lane-fort-hamiltom-pkwy-complete.shtml <a href="#fnref:2" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:3" role="doc-endnote">
<p>Maria Kaika, City of Flows: Modernity, Nature, and the City (New York: Routledge, 2005), 28. <a href="#fnref:3" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:4" role="doc-endnote">
<p>“Last year was the deadliest under Vision Zero. Here’s how Mayor Adams can save lives in 2022,” Transportation Alternatives, January 26, 2022. https://www.transalt.org/writing/last-year-was-the-deadliest-under-vision-zero-how-mayor-adams-can-save-lives-in-2022 <a href="#fnref:4" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:5" role="doc-endnote">
<p>Jake Offenhartz, “Bud Light Truck Driver Kills Delivery Worker Near Controversial New Bike Lane,” Gothamist, November 13, 2020. https://gothamist.com/news/bud-light-truck-driver-kills-delivery-worker-near-controversial-new-bike-lane <a href="#fnref:5" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:6" role="doc-endnote">
<p>New York City Bike Map, New York City Department of Transportation, 2021. https://www1.nyc.gov/html/dot/downloads/pdf/nyc-bike-map-2021.pdf <a href="#fnref:6" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:7" role="doc-endnote">
<p>Yingling Fan, Andrew Guthrie, and David Levinson. “Waiting time perceptions at transit stops and stations: Effects of basic amenities, gender, and security.” Transportation Research Part A: Policy and Practice, Vol. 88 (2016): 251-264. https://doi.org/10.1016/j.tra.2016.04.012. <a href="#fnref:7" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:8" role="doc-endnote">
<p>Greg Lindsey, “Bicycles, Gender, and Risk: Driver Behaviors When Passing Cyclists,” University of Minnesota Gender Policy Report, July 15, 2019. https://genderpolicyreport.umn.edu/bicycles-gender-and-risk/ <a href="#fnref:8" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:9" role="doc-endnote">
<p>“How Crescent Street Came to Be Safe for Cycling,” Transportation Alternatives, August 10, 2020. https://www.transalt.org/writing/how-crescent-street-came-to-be-safe-for-cycling <a href="#fnref:9" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:10" role="doc-endnote">
<p>Amy Zimmer, “Here Are City’s Top 15 Gentrifying Neighborhoods,” DNAInfo, May 9, 2016. https://www.dnainfo.com/new-york/20160509/crown-heights/here-are-citys-15-gentrifying-neighborhoods/ <a href="#fnref:10" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:11" role="doc-endnote">
<p>“Astoria Neighborhood Profile,” New York University Furman Center, https://furmancenter.org/neighborhoods/view/astoria <a href="#fnref:11" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:12" role="doc-endnote">
<p>Edwin DeJesus, “Op-Ed: I Love Biking Too, but the Crescent Street Bike Lane is a Disaster,” September 17 2021. https://licpost.com/op-ed-i-love-biking-too-but-the-crescent-street-bike-lane-is-a-disaster <a href="#fnref:12" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
<li id="fn:13" role="doc-endnote">
<p>Karen Chapple and Miriam Zuk. “Forewarned: The Use of Neighborhood Early Warning Systems for Gentrification and Displacement.” Cityscape 18, no. 3 (2016): 109–30. https://www.jstor.org/stable/26328275 <a href="#fnref:13" class="reversefootnote" role="doc-backlink">↩</a></p>
</li>
</ol>
</div>Alexa JakobIn which Queens (kind of) gets the moneyBenchmarking Josephson Junctions for Scalable Quantum Computing2022-05-27T00:00:00+00:002022-05-27T00:00:00+00:00https://wolframalexa.github.io/blog/2022/05/27/quantum-computing<p>For my senior capstone project at the Cooper Union, my classmates Tamar Bacalu, Mark Koszykowski, and I worked with New York University’s <a href="https://wp.nyu.edu/shabanilab/">Shabani Lab for Quantum Materials and Devices</a>. It was very exciting to be able to work with PhD students and postdocs in a new lab. In this blog post, I’ll give some background on the work and talk about our tentative results.</p>
<h2 id="motivation">Motivation</h2>
<p>The Shabani Lab’s research focuses primarily on the hardware involved in quantum computing, particularly semiconductor-superconductor junctions. One such junction is the Josephson Junction, which consists of two superconductors sandwiched around a semiconductor or insulator - this was our focus.</p>
<p>Quantum computing is becoming more widespread, and it has important applications in fields like cryptography, finance, drug discovery, and weather modeling. However, the hardware development process can be quite long, as the validation process is hard to scale. In order to measure quantum properties, we must bring a chip down to near-absolute zero. This takes days, uses a lot of energy, and is usually fairly expensive. It’s also hard to scale this for chips with many qubits because each pad must get its own lede attached at the beginning. However, there are existing methods in the semiconductor industry, where a chip is mounted onto a machine that programmatically probes the pads of each transistor, at room temperature. As a result, it would be helpful to determine some relationship between the properties of a qubit at absolute zero and at room temperature - that way, we could use existing methods to get a sense of the chip’s performance, and extensive validation at absolute zero would be a second step.</p>
<h2 id="background">Background</h2>
<p>There are two important properties for a Josephson Junction - critical current (Ic) and normal resistance (Rn). At absolute zero, when the junction is superconducting, the critical current is the most superconducting current a junction can support, and the normal resistance is the resistance of the junction after it stops superconducting (as calculated by finding the slope of the IV curve). A higher IcRn product is more desirable and signals a “better” junction.</p>
<p>There is some reason to believe that room temperature measurements can predict these two quantities - researchers with the Air Force found it to be true for Nb junctions in 2000 <a href="https://ieeexplore.ieee.org/document/913141">(link)</a>. Our material stack is different, so we expect to get different results.</p>
<p>The Shabani Lab has measured these devices at low temperature, so we simulated and measured them at room temperature and compared the results to existing data.</p>
<h2 id="full-paper-poster--presentation">Full Paper, Poster, & Presentation</h2>
<p>For more technical details, the paper, poster, and presentation can be found below.</p>
<h3 id="poster">Poster</h3>
<script async="" class="speakerdeck-embed" data-id="f6c22f8260bf49d69ee9570c452c7403" data-ratio="1.22840690978887" src="//speakerdeck.com/assets/embed.js"></script>
<h3 id="presentation">Presentation</h3>
<script async="" class="speakerdeck-embed" data-id="37f7d4231587413f88a4b2237658159d" data-ratio="1.77725118483412" src="//speakerdeck.com/assets/embed.js"></script>
<h3 id="paper">Paper</h3>
<script async="" class="speakerdeck-embed" data-id="c0fb7044e8b0452a8e8a3187bdfb8af6" data-ratio="0.772727272727273" src="//speakerdeck.com/assets/embed.js"></script>
<h2 id="results--learnings">Results & Learnings</h2>
<p>The results were, well, inconclusive. This is a hard problem to solve! As we note in the paper, many challenges prevented us from gathering the data we needed: lab access due to COVID-19, mismatch in data, and the physical impossibility of gathering certain results at room temperature. That said, we hope that we documented the process well enough that future students can further our work. Even though we were disappointed at this conclusion, it was important to accept the inconclusive results and not force a relationship that wasn’t there - that’s bad science. The lab was very supportive of our work and reminded us that no matter the outcome, the work we did furthered their goals by helping figure out new pathways to research.</p>
<p>I also learned a lot about experiment design and troubleshooting. A large part of research is not glamourous; you have to spend a lot of time making sure the peripherals work so that you can spend a little time on your main research question. You need to spend those three hours making sure you have the correct drivers installed because Labber (software that automates measurements) isn’t recognizing your instruments and fiddling with the pins on the probe station before every measurement to get a good electrical connection with the chip. You also need to be organized and systematic. Early on, a grad student, Mehdi, gave us a list of some common errors, so every time we would ask for his help, we’d make sure we ran through the checklist first, making sure it wasn’t a driver error, that our electrical connections were good, that everything was on, that the addresses were correct, that we had picked the correct chip and device, and so on.</p>
<p>I really enjoyed working on a research project over a long period of time, exploring different aspects of the problem, and learning about a new field that I didn’t have previous exposure to. Working with physicists also gave me more skills in collaborating with researchers in the pure sciences (as opposed to engineers).</p>
<h2 id="acknowledgments">Acknowledgments</h2>
<p>Again, many thanks to my group members Mark Koszykowski and Tamar Bacalu. They were joys to work with. I also want to thank Neveen Shlayan at the Cooper Union, who advised us, as well as Javad Shabani, Joe Yuan, Billy Strickland, Zhujun Huang, and Mehdi Hatefipour of New York University, who were indispensable in making this project work. I learned so much from them and I am thankful for them generously sharing their time, space, and knowledge with us.</p>Alexa JakobFor my senior capstone project at the Cooper Union, my classmates Tamar Bacalu, Mark Koszykowski, and I worked with New York University’s Shabani Lab for Quantum Materials and Devices. It was very exciting to be able to work with PhD students and postdocs in a new lab. In this blog post, I’ll give some background on the work and talk about our tentative results.How I Prepared for Electrical Engineering Interviews as a New Grad2022-01-10T00:00:00+00:002022-01-10T00:00:00+00:00https://wolframalexa.github.io/blog/2022/01/10/hardware-interviews<h2 id="my-experience">My Experience</h2>
<p>If you’re new here, hi! My name is Alexa, and I did a lot of electrical engineering interviews last fall. My background: I’m graduating with a Bachelors of Engineering in Electrical Engineering from the Cooper Union in Spring 2022. After graduation, I will join the electrical engineering team at Formlabs. In the past, I worked on board design at NVIDIA on the GPU Products Team for two summers.</p>
<p>I interviewed for similar positions to my past experience (board design and general EE) because I liked working on a broad range of topics, and wasn’t sure if or where I wanted to specialize. As a result, this guide may not be helpful for specific positions like RF and VLSI. Although I did have a few domain-specific interviews, my understanding is that the companies were looking for engineers with masters’ degrees for those roles, and I, as a senior in undergrad, didn’t fit that profile. It happens! As a caveat, I mostly interviewed with startups and smaller companies rather than Big Tech, so this guide may reflect broader/easier/harder/different questions than you might get from, say, Facebook.</p>
<h2 id="why-this-guide">Why This Guide</h2>
<p>When preparing for my interviews, I didn’t find a lot of information on how to prepare for EE generalist interviews. Software engineers have Cracking the Coding Interview, Leetcode, and more, but I haven’t been able to find similar resources for hardware. I think this is because hardware is simultaneously more broad and more niche than software engineering tends to be - there are fewer of us and interviews are less formulaic, so it doesn’t make sense to “grind” in the same way.</p>
<p>Finally, recently, a mentee asked what I had done to prepare for my hardware interviews, and I couldn’t point her to many resources. I figured writing up this guide would be helpful to others as well.</p>
<h2 id="the-basics">The Basics</h2>
<p><em>Know what’s on your resume.</em> If you say you can program in C, don’t be surprised when you’re asked to do it! This goes especially for past projects - you should be able to talk about the decisions and contributions you made in depth.</p>
<p><em>Explain your ideas clearly.</em> Your interviewer should understand the assumptions you made and the steps you took to get to a solution. This is just as important as being technically correct - engineers who can clearly explain their thought process are good colleagues and save the team time and energy.</p>
<p><em>Don’t neglect behavioral interviewing.</em> Being pleasant, curious, and organized makes you a good teammate. Prepare some stories about past school or internship projects that show how you work with others.</p>
<p><em>Research the company, team, and role.</em> This gives you an idea of who you’re talking to and what topics they could ask, and allows you to ask smart questions about the job. Some good tools for this include LinkedIn and Glassdoor. You can also ask people in your network for advice - one of my professors gave me some awesome advice in preparing for an interview.</p>
<h2 id="specific-topics">Specific Topics</h2>
<h3 id="circuit-analysis">Circuit Analysis</h3>
<p>Understand the basics of circuit analysis. Solve a circuit using KVL/KCL and equations for RLC. In school, you’re often content with finding solutions for every voltage and current in the circuit. You may not need to do that here - you can talk instead about tradeoffs with cost, power consumption, and precision.</p>
<p>[easy example: I like Question 1 <a href="https://kevinfronczak.com/blog/circuit-design-interview-questions-part-i">here</a>]</p>
<h3 id="electronics">Electronics</h3>
<p>Understand and apply BJT/FET rules, and how to use those devices to create basic circuits (namely amplifiers). You may also be asked about opamps.</p>
<p>[easy example: build an amplifier that takes a 10mV signal and amplifies it to 10V. What are some different ways you could do this, and what are the tradeoffs?]</p>
<h3 id="signal-processing">Signal Processing</h3>
<p>Understand what an LTI system is and why it is useful. Explain and apply the Nyquist sampling theorem and selecting from common filters.</p>
<p>[easy example: what happens when you sample below the Nyquist frequency?]</p>
<h3 id="digital-logic">Digital Logic</h3>
<p>Simple circuits (AND, NAND, OR, NOR, XOR), multiplexers, half/full-adder, flipflops, latches.</p>
<p>[easy example: design a NAND gate using 2:1 muxes]</p>
<h3 id="computer-architecture">Computer Architecture</h3>
<p>Understand generally how computers work: how instructions are handled, how memory is accessed, caching.</p>
<p>[easy example: what is pipelining, and how does it work?]</p>
<h3 id="programming">Programming</h3>
<p>Not all electrical engineers know how to program, but if you are one of the ones that does, that can be a huge advantage. In my interviews, I used C, for embedded applications, and Python, for everything else. Your code generally doesn’t need to compile, but you should understand and explain the choices you are making in the design process. At a minimum, I suggest understanding:</p>
<ul>
<li>how the language handles memory (malloc/free in C and static vs dynamic memory)</li>
<li>how to declare and use functions</li>
<li>using common built-in functions (like len() or list.append() in Python)</li>
<li>space and time complexity of the program you write</li>
<li>different datatypes and how they are handled</li>
<li>scope of variables (global/local) and pass-by-value vs pass-by-reference</li>
</ul>
<p>[easy example: divide an integer by 4 without using the ‘/’ operator]</p>
<p>Best of luck with your interviews!</p>Alexa JakobGirlboss Energy OnlyEthics Explorer: A Web App for Thinking About Ethical Engineering Careers2022-01-02T00:00:00+00:002022-01-02T00:00:00+00:00https://wolframalexa.github.io/blog/2022/01/02/ethics-explorer<h1 id="vision">Vision</h1>
<p>This semester, I decided to do an independent study about engineering ethics, a topic that matters to me and that I feel is not centred enough in our curriculum. Because my post-grad job search was top of mind, I was motivated to explore employment ethics. I wanted to know how people determine which jobs to take and how they balance ethics with other factors in that decision. Of course, we live in the real world, and we don’t always make the most ethical choice all the time. There are other factors to consider when choosing a job, like one’s skills and interests, employers who are hiring, compensation, location, and team and company culture. That said, I hoped I could shine some light on ethical aspects of career choices and plant seeds of ethical inquiry without being prescriptive or overbearing.</p>
<p>As a result, I built <a href="https://ethics-explorer.glitch.me/">Ethics Explorer</a>, where students can select a major, an industry, and an ethical area of interest and receive information and further reading about relevant challenges.</p>
<h1 id="about-the-tool">About the Tool</h1>
<h2 id="philosophical-design">Philosophical Design</h2>
<p>Initially, I was going to suggest specific careers, but that didn’t feel right - who am I to prescribe a career to anyone? And besides, any point system or recommendation I would make would be inherently flawed. Instead, because I wanted to spark thinking, I chose to present each issue as a “challenge” - which users would have to take up and act on, in their own way. Every profession and every industry has its challenges, but every system, no matter how big, is perpetuated by humans. I hope we can take responsibility for our own actions and consider how we might act to make the system more just.</p>
<p>The same career can be ethical or unethical depending on who you ask, so the application matters more than the skills. I chose to focus on six specific areas, undoubtedly leaving out others just as important: equity, labor, corporate citizenship, privacy, the environment, and safety.</p>
<h2 id="technical-design">Technical Design</h2>
<p>This is the first webapp I’ve ever built, which was a big challenge - more on that later! The app is hosted on <a href="https://glitch.com/">Glitch</a>, a free web hosting platform. I used SQLite3 and Node for the backend, and JavaScript, HTML, and CSS for the frontend.</p>
<p>I created the database myself, according to the following schema:</p>
<center><img src="/assets/images/er-diagram.png" alt="ER Diagram" /></center>
<p>I chose to implement this fairly simply. I didn’t use an ORM for the database and opted to write SQL queries in JavaScript instead. The user can only read from the database (no CRUD - create, read, update, delete - functionality). There is only one page, rather than having multiple tabs. Some of this is functionality I’d like to add in the future.</p>
<h1 id="learning-web-development">Learning Web Development</h1>
<p>I knew I would have to code a web app for the Databases final project, but I didn’t know how to do it. Even concepts like client and server side (frontend/backend) were foreign to me - I essentially learned webdev basics in a week.</p>
<p>My biggest hurdle was understanding concepts between client and server side. Because I can code in Python and other languages, I could follow the coding logic, but the bifurcation between front and back end was confusing to me. I asked for a lot of help, mostly from my friend <a href="https://riccc.cc/">Ricky Yurewitch</a>, an art student at Cooper whose practice includes a lot of digital art. I made sure I was asking well-researched questions to understand, rather than to fix my specific problem. I’m always cognizant of taking up <em>too much</em> of someone else’s time, so I try to pack in concept-level understanding so that I can find quality information by myself later on.</p>
<p>A lot of people say that software engineering is basically being good at websearching, and they’re kind of right. I’m grateful for the methodical way my engineering degree has taught me to solve problems. In this case, after obtaining the conceptual knowledge, I knew what I was trying to accomplish, could search for proper information (eg “how to query from database in javascript”), and select high-quality information from the results.</p>
<p>Initially, my project was based on a Cooper Union art professor’s CRUD database project, but understanding his design decisions was difficult. My data was also structured differently than his, and we had different applications. I spent a lot of time trying to figure out his schema initialization, but it ended up being easier to just start from scratch. It’s important knowing when to quit, and not give into sunk cost.</p>
<h1 id="difficulties">Difficulties</h1>
<p>First, how do you reduce such a complex problem to a database? This is obviously not the end-all be-all to engineering ethics, nor should it be. I fear that people will take the recommendation as gospel, and not as a starting point for individual reflection. I’ve made it clear through the text that it wasn’t my intention, but can’t do much else. Even so, can a tool like this be helpful, as limited as it is? I hope.</p>
<p>Technically, data cleanliness was a challenge. The United States Bureau of Labor Statistics data does not agree on how many workers there are in the United States in the tables for race/gender and union density. As a result, I focused on percentages rather than raw numbers. I assembled the database, so there might be some errors.</p>
<p>I chose to make the database in SQL for simplicity, but, thinking back, a NoSQL database like MongoDB may have been more extensible, with more data and links. For example, there are three columns that hold URLs, and many of the entries are NULL. If I wanted to add another URL to a record that already has three, I would have to create a new column for every record.</p>
<p>Implementing the database was challenging. Like Python, you need to specify SQLite 2 or 3, so my database did not “exist” for a while because it was the wrong version. Using SQL rather than an ORM avoided the extra complexity and upkeep that comes with using one, and I didn’t need to initialize the database schema. I certainly could have made different choices for implementing this database.</p>
<h1 id="future-work">Future Work</h1>
<p>I have spoken with the Center for Career Development and with instructors for EID101 (Engineering Design & Problem Solving), to refine the tool according to their needs. I hope they can use the tool to spark discussions with students about ethics.</p>
<p>As for the tool itself, I’d love to add pedagogical value by prompting users to reflect on the ethical challenges of a particular field before showing them the results. It would be even better to also add suggestions, and use both of these fields to further refine the tool.</p>
<p>Adding more and higher-quality data is also important. Currently, the industries are matched with industries from BLS data, but they aren’t standardized across BLS tables. It would be good to use better-defined industries and data and include more engineering majors.</p>
<p>Much further down the line, I’d like to provide space for reflection and discussion around engineering and its role in society. This could involve functionality where people share information about their project or employers, discuss case studies, and receive support for navigating their own situations.</p>
<h1 id="final-words">Final Words</h1>
<p>The tool can be found here: <a href="https://ethics-explorer.glitch.me/">Ethics Explorer</a>
The code can be found here: <a href="https://github.com/wolframalexa/ethics-explorer">GitHub Link</a></p>
<p>Please <a href="mailto:alexajakob@tutanota.com">email me</a> or send a pull request if you see something you’d like fixed or have questions!</p>Alexa JakobMy very first web app!Designing an Open-Source Baseboard to Empower Hardware Makers2021-12-23T00:00:00+00:002021-12-23T00:00:00+00:00https://wolframalexa.github.io/blog/2021/12/23/baseboard-slides<p>I recently gave a talk on the design process for My Jetson Nano Baseboard, my intern project over the course of my two summers working at NVIDIA. I’ve included the slides below. I chose to discuss this project because it is the most large-scale project I’ve worked on, and I was able to develop skills from the beginning to the end of the product design process.</p>
<h2 id="abstract">Abstract</h2>
<p>NVIDIA’s <a href="https://www.nvidia.com/en-us/autonomous-machines/embedded-systems/jetson-nano/">Jetson Nano</a> is a popular product for developers who want to get started with AI on a small scale, and it is making AI learning accessible at a low price point.</p>
<p>Makers are excited to take the device’s capabilities further, but few hardware design resources exist. This project provides an open-source example board with accessible cost and complexity for makers to kickstart their own projects. Over the course of my internships, I gathered requirements, designed a schematic and layout, validated, debugged, and released this board on GitHub. The following slides are a deep dive into some of the technical challenges faced and design decisions made.</p>
<h2 id="slides">Slides</h2>
<script async="" class="speakerdeck-embed" data-id="ab57303cdfc04f3a9b6666e06796e510" data-ratio="1.77777777777778" src="//speakerdeck.com/assets/embed.js"></script>Alexa JakobI recently gave a talk on the design process for My Jetson Nano Baseboard, my intern project over the course of my two summers working at NVIDIA. I’ve included the slides below. I chose to discuss this project because it is the most large-scale project I’ve worked on, and I was able to develop skills from the beginning to the end of the product design process.