In 2016 I finished a PhD in neuroscience and made the move to tech. Since then, I’ve worked as a data scientist and product manager in several companies. I’ve also spent a lot of time screening and interviewing candidates for data science roles.
I frequently receive questions about how to execute this transition successfully. This post is an attempt to summarize my advice. It draws upon help I’ve received from lots of other people who have also made this transition. …
Disclaimer: this post concerns two things about which I don’t know very much: economics and climate policy. If you’d rather read something I’m supremely qualified to talk about, try this post on rookie errors I’ve made in my machine learning career.
Jimmy Wales, the founder of Wikipedia, created the most valuable product of the 21st century and gave it away for free. I was therefore a little surprised to read that he’s a fan of Friedrich Hayek, the Austrian economist commonly associated with neoliberalism and the Chicago school of economics. In particular, a 1945 piece entitled The Use Of Knowledge…
Olya is the CEO of Frost Methane, a startup which “flares” (burns) methane. Flaring reduces methane’s warming potential by 28x by turning it into methane’s more famous but less-warming cousin, CO2. We discussed her journey in climate tech, selecting high-leverage problems, and how newcomers in climate tech should give her a call.
The biggest AI news of 2020 so far is the success of OpenAI’s monstrous new language model, GPT-3. In this post, I’m going to quickly summarize why GPT-3 has caused such a splash, before highlighting 3 consequences for individuals and companies building things with AI.
Why are people excited about GPT-3? Here’s why, in 3 tweets:
There are already lots of summary posts about GPT-3, so I won’t rehash them here.
For a great introduction to how the model works, check out this visual guide from the (reliably excellent) Jay Alammar. …
This time last year, my co-founder Pat and I had a full schedule, and most of it revolved around work. Between the corporate frisbee league, the employer-sponsored HIIT classes, and the weekly company pick-up soccer game, we were seeing a lot of our colleagues. Add in after work drinks a couple of times a month and we were spending upwards of 8 hours a week socializing with our team.
Fast forward to Summer 2020, and the social fabric of our work life had been destroyed completely. All sports had been cancelled. Everybody had come to dread the weekly work drinks…
During the pandemic, Statistics Canada has been compiling data on how peoples’ work habits have shifted. We’ve been digging into this data, and in this post we’re going to share some insights and discuss how we think these trends will develop throughout the rest of 2020.
Prior to February 1st, only 10% of workers spent more than half their time working remotely, but as of March 31st, this was up to 34% — an increase of 240%.
Crunching the numbers, this corresponds to:
9 extra hours a week worked remotely on average.
140 million extra hours per week across Canada.
Over the past three years, I’ve spent >50% of my time thinking about what the applied research teams I’ve been part of should be building, and how. This post is about some of the challenges I’ve faced helping to organize applied machine learning research in a hyper-growth setting.
These observations are subjective and overfit to my personal experience; if you’re leading teams at Google Brain, this blogpost is probably not for you. …
At the end of March, Quebec’s Premier Francois Legault’s message to the people was clear: stay inside. But he didn’t stop there, offering a tongue-in-cheek remedy to the anxiety of lockdown: “a glass of wine may help”.
Since SAQ (Quebec’s central wine stores) remains open as an essential business, we were curious to see whether people were taking Legault’s advice. We’ve been scraping data from the SAQ website since the start of April to take an inventory of the stock and understand whether more people are turning to Cote du Rhone or Chardonnay to take the edge off their isolation.
Update 19/03/20: thanks to lots of fantastic feedback, we’ve now released V2 of the tool, which brings historical case data per day for your country, better graphs, and a breakdown of forecast mortality by age! It’s still at https://corona-calculator.herokuapp.com ✌️
Many people have been persuaded by recent developments that we should take coronavirus very seriously indeed. Posts such as this one underline how important isolation has been as a tactic to control the spread of Coronavirus. You’ve probably seen scary curves of number of infected individuals:
This is a written version of a talk I gave at McGill University in January 2019. Subsequently, Andrej Karpathy wrote an excellent post, somewhat more technical and deep-learning specific but with overlapping content — check that out here.
You can track your deepening expertise of a topic by the quality and quantity of the mistakes you make. This post catalogues some of the more repetitive errors that I and the people around me have made in machine learning, in the hope of accelerating your progress out of the shallow waters of inexpertise and into the deep dark waters where really…
Product manager & data scientist. Writing about AI, building things, and climate change.