How I use LLMs to learn new subjects - It’s good at well know facts, things that are mainstream or so niche that there is only one answer: “you should avoid asking the model for concrete details that don’t really matter, and if the model gives you details like that you shouldn’t trust them. But when the model is speaking about facts that do matter - facts that are load-bearing for many other things the model knows about - you can be relatively confident that you’re getting correct information.” // Also, nicely put: “models are not minds; instead they are role-playing a particular kind of helpful character. They prioritize consistent role-play above almost all other concerns."
Posts in "links"
The more senior engineers get, the more results matter - “as you become more senior, you’re increasingly graded on results. Interns are graded on effort."
Kubernetes Complexity Realigns Platform Engineering Strategy - Turns out computer are always difficult.
Where Technology Executives Will Be Investing In 2026 - Looks like lots of datacenter buildout.
Code was the least interesting part of my multi-agent app, and here’s what that means to me - Less time coding means more time designing and product managing.
Java licensing snafus cost millions, drive developers to open source - ”One-quarter of respondents said their organization spent between $50,000 and $100,000 resolving software non-compliance issues, while licensing issues cost 17% of surveyed organizations up to $1 million.”
Beyond London Summit 2025: Decoding Google Cloud’s Scale Narrative - Summary of stated Google cloud strategy, from Bola.
AI industry’s size obsession is killing ROI, engineer argues - Doing enterprises AI isn’t free in terms of time, money, opportunity cost, and risk. You need target things that have a pay off, an ROI. I hope that this will mean doing more, not cutting costs. That is growth, not firing people.