#development / Page 4

In training agents, we toss the whole run if the final outcome is imperfect, missing out on valuable info. To fix this, we developed Step Rejection Fine-Tuning.

GitHub has introduced the GitHub Copilot app, a desktop control centre for agent-native development that aims to keep engineers in charge while AI agents handle more coding work. Mario Rodriguez writes on the GitHub blog that the recent wave of ...

The global alliance helps organizations modernize software delivery, secure their supply chain, and bring AI into development workflows.

Jeff had decades in software but mediocre Python and no map of AI. Six weeks later: an agent with three interfaces, ~250 tests, and a real mental model.

Browser automation lives or dies by where it sits in the development workflow. The exact same set of automated checks can be either the most valuable safety net a team has or a despised source of friction, and the difference is almost entirely about ...

Four credible 128GB-class boxes, four very different price points. We synthesise what practitioners with the hardware on their desks are actually reporting.

VibeThinker-3B is WeiboAI's MIT-licensed 3B reasoning model built on Qwen2.5-Coder-3B. We unpack the viral 'Opus 4.5 performance' claim with the actual HF benchmarks.

GLM-5.2 is Z.ai's frontier open-weights model -- coding and agentic upgrades over GLM-5.1, live on OpenRouter, picked up by Nous Hermes Agent within days. Our hands-on guide.

JIT Delayed, Sandboxes, OpenRouter, and More ...

On July 9 (9am-1pm Pacific), I'll be teaching a 4hour live workshop for O'Reilly: Building Data Apps with Streamlit and Copilot. This is the second time I've run this workshop, a ...