Scaling Data, Regression-free ML Model Migration, The True Meaning of Technical Debt, and More
Featured posts from Dropbox, Doordash, and More
I hope you and your family are doing well and that you can find a new rhythm in this difficult situation. As usual, I've compiled a list of my favorite discoveries from the previous month. I hope the links from this edition inspire you. If you missed the last one, you could catch up by clicking here.
If you like this newsletter, please forward it to others who might be interested.
This week I came across this article about the true meaning of technical debt and it did change my perspective about technical debts. It’s common to think of tech debt as something that piles up as a result of bad coding practices which is far from being true. @lucaronin captured it well in this post. Tech debt is a disagreement between business needs and how the software has been written. Understanding the true meaning of tech debt can help devise a strategy to limit it.
Migration from one system to another is fairly common. Despite the fact that it was not a machine learning model, I spent some time in the last month migrating one of our key features to a new system. As your organization's and business's needs evolve, you'll discover that a system that worked well for a few years may no longer meet your requirements for speed or agility. However, how does one move a machine learning model without introducing regression? Successful ML system migrations imply that the new service performs well, if not better than, the old one. This article discusses some best practices for avoiding regressions when migrating ML models.
How do you view data? A set of tools to buy? Or a team to hire? “One common mistake that some organizations make is viewing data as a team to hire or set of tools to implement rather than as a strategic lever for growth.” In this post, @crystalwidjaja, A former SVP of Growth and Business Intelligence at Gojek explained why viewing data as a team or set of tools to implement is wrong and how to begin approaching data as a strategic lever for growth.
“Like a house built to withstand the seasons, a successful and sustainable analytics project must start with a firm foundation, a purposeful plan, a seasoned team, and the right tools and materials.” I came across this article and I’m happy to share it. It will help you understand the right questions to ask for successful analytic projects.
AI and Machine Learning and Deep Learning are often used sometimes interchangeably. But what is the difference?. Many people don’t always seem to understand the meaning of these high-frequency words and the relationship behind them.
This article explains the meaning of these words in the most straightforward language and clarifies the relationship between them.
“One of the most significant transitions for an engineer’s career is when they break dependence on senior team members and start to contribute to the team’s decision-making process actively. Once this transition is complete, the engineer contributes to the team at levels far more significant than the code they can sling. They become a more productive team member and help unlock team potential by freeing up others’ time and effort.” How do you progress to become an influential engineer? This article will help you.
Last year, I read Your First 90 Days by Micheal Watkins. It’s a good read for anyone transitioning to a new role at a new organization. @lethain, the CTO of Calm, outlines what to do in your first 90 days as CTO or VP Engineering.
“Leadership is what converts a group of people into a team. As a Tech Lead, your maximum potential effectiveness is now multiplied by the size of your team. Your full potential can only be reached after your team has reached theirs.” This post contains some valuable lessons for engineers transitioning from managing technology to leading and managing people.
“Pull requests with blocking reviews (sometimes mandatory) are widespread in our industry. A lot of developers believe pushing straight to the main branch should be prohibited. Sometimes it’s unavoidable (in a low-trust environment), but often people work with PRs just because everyone else does. And nobody ever got fired for it. But what are the costs of working in such a style? And what are the alternatives?”
Photos are among the most common types of files on Dropbox, but searching for them by filename is even more difficult than it is for text-based files.– In this post, the Dropbox team describes the core idea behind their image content search method, based on techniques from machine learning, and how they built a performant implementation on Dropbox’s existing search infrastructure.