Jeremy

did you know cloudflare encrypts the internet using a wall of lava lamps? 100 of them sit in their SF lobby. a camera films the chaos, the images get hashed, and the result seeds the cryptographic keys for ~20% of global web traffic. the london office uses a wall of double pendulums. the austin office uses hanging rainbow mobiles. the singapore office uses a chunk of uranium in a glass jar. just in case the lava lamps stop being weird enough.

AI agents are only as good as their context. 🤖 Give your agents secure, real-time access to your Datadog telemetry, monitors, and dashboards with the new Datadog MCP Server! Watch the latest episode of This Month in Datadog to learn how: bit.ly/4wjFqy0

It isn't unexpected that the focus of the Bun Rust rewrite is on the anti-Zig side more than anything, since the internet loves to hate. What is unexpected and unfortunate is that leadership within Bun hasn't tried to steer the conversation away from that at all. There are so many positive and interesting takeaways from this and I'm not really seeing any of them pushed as the primary message. A positive thing that hasn't been talked about at all is how far Bun came thanks to Zig. And even if you dump it now, its meaningful for how good Zig was to even build a product to this point and impact by any metric. I would've loved to see anyone in leadership say this. On the interesting side is how fungible programming languages are nowadays. Programming languages used to be LOCK IN, and they're increasingly not so. You think the Bun rewrite in Rust is good for Rust? Bun has shown they can be in probably any language they want in roughly a week or two. Rust is expendable. Its useful until its not then it can be thrown out. That's interesting! There's been a lot of talk about memory safety and no doubt Rust provides more guarantees than Zig. But I'd love to see a better analysis of why Bun in particular suffered so much rather than take the language-blame path. How could engineering as a practice been more rigorous to prevent this? What were the largest sources of crashes other programs should watch out for? How does Rust prevent them? How could Zig theoretically prevent them? That's interesting. I know the official blog post hasn't come out yet from Bun. But they're smart enough to know that that PR would stir up controversy the moment it opened, or they should've been. And plenty in the company have been tweeting and writing about it. Its somewhat telling to me in various dimensions what they chose to talk about first. I tend to think I'm pretty good at corporate PR/comms (especially when it comes to developer audiences) and I think appealing to the negative is never the right long term strategy; it does work to get short term eyes though.

Toto 2.0 is here: Datadog AI's 5 open-weights forecasting models (4m-2.5B params) finally make scaling work for time series forecasting! #1 on BOOM, GIFT-Eval, and TIME. Weights/code Apache 2.0. 🔗 Read the blog post for more details: bit.ly/4tCFvKL

🎬 April’s This Month in Datadog is live, and we are showing you how teams are using AI to move faster with better context across their systems. Want to see what’s next? Join us at DASH for an immersive look at the latest in AI-powered observability and security. Secure your spot: dash.datadoghq.com/?utm_source=tw…

Wyndo answers your weather questions based on input from @OpenWeatherMap @open_meteo @NWS and is hosted on @Railway and @Cloudflare Pages. Weather apps tell you it'll rain. Wyndo tells you when to go. wyndo.app

I can feel LLMs lightly eroding my brain's willingness to focus on anything hard for more than 10 minutes, even things that I enjoy, and this is maybe the most scared I've ever been of a consumer technology.

NEW: Wake up to today's answer. Set a 7am run routine; get a yes/no in your inbox or on the dashboard. Logged in users only! wyndo.app/welcome?step=r…

NEW: Send a friend "should I run now?" as a live link. They get a real answer for where they are, not where you are. wyndo.app/run-now

New on wyndo.app: every recommendation comes with the why in one sentence. Engine decides deterministically. AI narrates. Can't hallucinate the weather, just translates the trace to plain English.

Bits AI Security Analyst, Bits AI Dev Agent, Bits Assistant, Spotlights on Datadog MCP and Datadog Experiments, and more… This Month in @datadoghq youtu.be/RoKS5wByv8U

The question: should i run now? weather apps: 78°, 60% chance of rain today. Wyndo: you've got 47 minutes before the storm. wyndo.app

Is "markdown cool again", or was it cool the entire time?

we literally have no idea how to make software anymore

unfortunately i need to stop playing with how cool the technology can be and ship practical things people can use

New: shareable URLs on Wyndo. Drop wyndo.app/run-now (or /walk-now, /bike-now, /dine-now, etc.) in a chat. Your friend lands on a real answer for their location, right now. Know when to go.

A profound error that many experienced product people make is to fall into the habit of thinking & speaking at the level of clever proxies (frameworks, industry jargon, corporate buzzwords) rather than seeing the basic facts of the customer situation & identifying what matters.

doing some prototyping of a plugin system with gpt 5.5 and it's reinforcing just how i don't see the code going away yet in this screenshot alone, the issues are: - it made plugin based data storage part of the core sdk rather than each plugin defining it's own data store - it introduced the concept of an 'actorId' because the cloud version needs to know who ran it but the local version doesn't need this - because of this it made 'actorId' nullable which in the cloud app could lead to bugs where you don't know who ran what which imo is a critical bug - it didn't strongly type "execution:started" i had to tell it to make it infer the input params from the string - same thing with collections - i cleaned this up already but it made the local plugin dependent on the cloud plugin that adds metadata about who ran it - this is a nit but you don't need to do .collections, you could just do .executions and have the same strong typing without needing to provide a cast - it added priority and after in without me telling it to which i need to understand better what the goals of those are but are likely wrong this was generated after a decently long chat on plugin systems in other codebases for it to make an MVP if i went and built my product on this foundation, it'd be like 2-3x the amount of code i have to maintain in the long run and would limit what it'd be able to do the counter point to that is if i just embraced it i could go ship a plugin system which would make my users happy immediately and just fix the architecture later i dunno maybe we get to a world where you can stop looking at the code but we're definitely not there yet they're in this weird state where this code looks incredibly plausible but has all of these hidden issues, but, it also does do the job that it needs to honestly i truly have no idea how to use these tools yet or what software engineering is going to look like - maybe none of these are problems or maybe they matter more now than ever

One takeaway for me from my recent "This Month in Datadog" discussion with the @datadoghq co-founders. It's time to question long held beliefs and habits. Many of them are still correct, and worth defending. Others aren't. Some are now liabilities. youtube.com/watch?v=RRl9KM…

My number one piece of advice to engineers in 2026 is to talk to customers. At any company size, if you’re an engineer, you are missing out on massive wins if you’re not talking to customers. You miss out on: * Information from what they want * Networking and relationship building * Empathy and urgency from understanding their challenges * The authority that comes with knowing customers On the last bullet - especially in big companies you’d be shocked at how powerful you are if you can say “customers I spoke with want X” with conviction.