Keep up with tech in 5 minutes
Get the free daily email with summaries of the most interesting stories in startups 🚀, tech 📱, and programming 💻!
Join 1,600,000 readers from companies like Anthropic, OpenAI, and more for one daily email
Does a URL in a prompt steer an LLM's output toward its content? (22 minute read)
When a URL and its content make it into the training data, the bare URL works as a key into the weights
Jun 12 | Blog
I Tried to Build a Context Layer for My Agent in a Weekend. Reader, I Did Not Build a Context Layer for My Agent in a Weekend.
A "simple" weekend project turns into real infrastructure, and why agent context deserves a boring, reliable foundation.
Sponsored
Jul 06 | Data
GitBiased (Tool)
GitBiased gives engineering teams a customizable dashboard for GitHub activity, bringing PRs, CI checks, issues, DORA metrics, deployments, and weekly summaries into one shared view.
Jul 03 | AI
Autoresearch, Claude, and Constrained Optimization (13 minute read)
Many folks claim to be able to use AI to do the work of dozens of people. This researcher used autoresearch to work on a problem where the path from unknown to success is a clear, gradient optimization. The researcher found the auto-research/loop style of work makes sense for problems with a robust, measurable, and well-constrained metric to optimize. However, finding a problem with these factors is often tricky.
Jul 03 | Founders
Are AI Employees More Expensive Than Humans? (8 minute read)
Token prices fell from $60 per million to $0.60 in three years, and AI bills still exploded. A four-person startup ran up a $113,000 monthly bill from a single provider. Uber torched its entire 2026 AI budget in four months. Consumption is the reason. An agent re-reads context and checks its own work, burning 60 to 140 times the tokens of one reply. Track cost per completed task, not per seat, and keep the model layer swappable, because cheaper tokens do not mean a cheaper bill.
