The Pulse #92: GitHub’s AI-assisted developer workflow vision
Also: Google laying off engineers as their business is booming; a good period for startup fundraising and IPOs; and how WorkOS acquired devtools startup Warrant
Programming note: next week, I’ll be on spring break, spending time in Florida, visiting my in-laws. This will mean no The Pulse on Thursday: but there will still be a Tuesday article, as usual. Thank you for your understanding and support!
The Pulse is a series covering insights, patterns, and trends within Big Tech and startups. Notice an interesting event or trend? Send me a message.
Today, we cover:
Industry pulse. Cloudflare handles an extra 12M req/sec with no issues, how does being a PM at Google or Meta compare, and are nearly 1 in 10 of all developers worldwide using GitHub Copilot? And more.
GitHub’s AI-assisted developer workflow vision: Copilot Workspace. Several startups are aiming to build an AI coding tool that goes beyond GitHub Copilot. With exceptionally fast execution, GitHub seems to have beaten them to it. It’s called GitHub Copilot Workspace, and doesn’t aim to replace devs. A review of this developer-driven, AI-assisted workflow, including insider details from the GitHub team.
Google lays off devs while business booms. Google made layoffs in its Python and Flutter platform teams, right as the company’s stock price hit an all-time high.
Good time for tech scaleups fundraising and IPOs. It’s been a while since we’ve had so much positive news on tech IPOs, and late and early-stage fundraising. Hopefully, this trend continues!
Platform scaleup buys feature startup: WorkOS acquires Warrant. Warrant is an innovative fine-grained access control seed-stage devtools startup, built on the principles of Google’s authorization system. In what looks like a win-win outcome, WorkOS has acquired it, intending to keep intact the team, product, and all existing customers. Details on this acquisition from the founder and CEO of WorkOS.
1. Industry pulse
Cloudflare doesn’t notice 12M extra req/sec
Cloudflare engineering director Benjamin Yule shared an interesting non-event:
I have so many questions as 12.5 million requests per second is a huge load! A few thousand requests per second is usually considered high, and handling a load in the millions surely requires thousands of servers (if not more!) to be spun up to handle it.
How much load a single server can handle depends on how long a request takes to handle, how much memory a request uses, and then calculating whether a request is limited in CPU or memory. Whichever resource (CPU capacity or memory) runs out first determines a server’s maximum load. You can push things pretty far with clever workarounds; it’s how the WhatsApp team served 2 million requests per second on a server with 24 cores and 100GB RAM, back in 2012!
In this case, Cloudflare soaked up the load by using Cloudflare Workers, a serverless application platform, which spun up 40,000 CPU cores (an average of 320 requests/second per CPU core.) The tweet says the Cloudflare Workers team didn’t even notice the increase, which suggests they could be handling 50-100x more load on the service! If we assume a 100x load, that could mean operating 4 million CPU cores or more simultaneously, just for Workers. By comparison, we previously covered travel booking service Agoda operating a total of 300,000 physical cores and 600,000 virtual ones. It’s likely Cloudflare is operating more than 10x this number.