Haggai Nuchi

I’m a Software Engineer and mathematician based in the Bay Area. My interests are in machine perception, multi-object tracking, machine learning applications to audio, and computational geometry (among other things).

I’m currently looking for opportunities for remote software engineering part-time work. I have a broad range of interests, but some examples of things I’d be excited to work on are robotics, sensor fusion, applications of machine learning to medicine. I enjoy working on novel technical problems that no one knows how to solve. Send me an email at hello@haggainuchi.com. See below for some highlighted work experience. For the keyword searchers, I have expertise in Python, C++, sensor fusion, robotics, Tensorflow, macOS, Linux, bash, geometry, mathematics.

Current activity

I am currently seeking part-time employment or consulting work.

Past activity

In November and December 2020, I did pro bono work tracking and counting Daphnia magna specimens in tanks of water from video. You can see the code and documentation here.

I was a software engineer at Airbnb from October 2018 to October 2020. I did machine learning for search ranking at first, and then moved to working on the machine learning platform (tools/ops platform for ML engineers at Airbnb). The latter project’s github page will eventually become public but is (as of this writing) still private. My contributions there were under my work github profile.

In Spring 2018 I shared my project translating human sounds to dog barks entirely in the browser. It works by compiling a Tensorflow model to WebAssembly. I’ve shared some of the details of how it works in that link, and you can try it out, all in the browser!

In Fall 2017 I did some mathematical thinking on the Wobbly Table problem and the Square Peg conjecture. I was introduced to the Wobbly Table problem by some former colleagues at Cruise Automation and it got my attention. I hadn’t done math in a few years and was excited to return to it after leaving my job at Cruise.

From Spring 2016 to Summer 2017, I was a software engineer at Cruise Automation working on autonomous driving. I was tech lead for the Prediction team (writing algorithms and code that runs on the autonomous cars and predicts where other drivers/cyclists/pedestrians will go, supervising interns, coordinating with other teams, making sure the software met latency requirements, etc), and also worked on multi-object tracking and sensor fusion.

Prior to working as a software engineer, I got my PhD from UPenn and worked as a postdoc at the University of Toronto, in mathematics. My research focus was in differential geometry, in particular around Hopf fibrations and related objects.

Contact

I’d be happy to hear from you if anything here catches your interest! If you check out the stuff I’ve written and you like it, send me a note at hello@haggainuchi.com. Thanks to a recent reader who read some of my work on the Square Peg conjecture and who wrote to me to inform me that there has been some progress made in the past few years, with some researchers solving the smooth rectangular peg problem!