A developer's analysis of roughly 4,300 real coding-agent sessions (Claude Code and Codex, 43 developers, ~350,000 LLM steps) found that the median LLM step re-sends about 119,000 cached prefix tokens…
Augment Code benchmarked ten open-source AI code review tools against a 450,000-file Python, TypeScript, Java and Go monorepo, finding that self-hosting costs roughly $4,100 to $9,100 per month at any…
Inception released Mercury 2.5, which it calls the most capable diffusion LLM on the market and the largest it has ever trained, reporting a 40% intelligence gain over Mercury 2 at 1,107 tokens per se…
Inception AI released Mercury 2.5, its most capable production model, claiming a 40% increase in intelligence over Mercury 2 with speeds of 1,107 tokens per second and a context window of 260K tokens,…
GitMir claims to outperform Augment Code, Unblocked, and Sourcegraph Cody in answer quality, achieving 92% accuracy at $0.07 per answer, according to its own benchmark. The product, which integrates v…
An agentic development environment (ADE) is a workspace for directing multiple AI coding agents simultaneously, providing isolated workspaces, activity visibility, persistent task/plan storage, and fi…
A vendor-agnostic framework from DORA and other sources shows that AI coding tools should be judged by delivery outcomes such as change failure rate and time to restore service, not by vendor benchmar…
Beth Andres-Beck, a software engineer and congressional candidate in Massachusetts' 6th district, argues that the real superpower of engineers is understanding why people do what they do, and that met…
A multi-agent coding workspace is reliable only when agents work on isolated, spec-scoped tasks because six coordination patterns prevent file collisions, duplicated implementations, and semantic drif…
The term 'Agent Development Environment (ADE)' has two distinct meanings in 2026: Letta's original sense of a platform to build and debug agents themselves, and a newer sense, popularized by Orca, of …
Jessica Kerr argues that AI has split the programmer's job, commoditizing hand-crafted coding while leaving the harder human tasks of understanding what to build, proving it works, and stewarding a li…
OpenAI's 'Practical Guide to Building Agents' has sparked a debate with LangChain CEO Harrison Chase, who called it 'misguided' and published a detailed rebuttal. The conflict highlights a core tensio…
Randy Shoup abandoned a legal career after a summer internship watching inventors create while he was relegated to note-taking, returning to Oracle to pursue engineering. He and Kent discuss the innat…
OpenAI has introduced Codex pets, optional animated companions for the Codex app that appear in a persistent overlay to show task progress. Users can create custom pets by installing the hatch-pet ski…