Topic
Everything filed under AI Agents, newest first.
RSS · JSON · All topics
Linear shares data on how its code review changed as agents wrote more code: every PR is now agent-reviewed first and authors can self-approve, so 53% of merged PRs in October 2026 had no second human reviewer, with mandatory human review kept for sensitive areas.
10 min · 2,322 words
Introducing Islo: the safe way to run enterprise software factories
Islo CPO Adam Gold launches Islo enterprise software factories, backed by $15M from Insight Partners, arguing that when code becomes nearly free an engineering org's job is producing trust at scale through factory lines, isolated machines and model-independent evals.
10 min · 2,208 words
Introducing FrontierHarness Eval v1.1
Runta and Baseten ran 22 coding-agent harnesses with 5 models on the same 62 tasks, 4 times each (27,280 runs, 83B tokens), to test whether switching models should mean switching harnesses, and found 64% of inference spend went to runs that failed.
11 min · 2,500 words
Windows 11 Is Collecting Data on Which Coding Agents and Developer Tools You Use
Reverse-engineering MissionControlAggregator.dll shows Windows 11 records which coding agents (codex, claude, copilot, cursor) and developer tools you run, stores it locally and reports it via optional diagnostic data, which is on by default. Includes how to opt out.
7 min · 1,527 words
SWE-CoWork: Benchmarking Coding Agents as Digital Employees in Organizations
TokenWave AI's SWE-CoWork benchmark drops coding agents into simulated engineering orgs with 26 colleagues across 101 real repositories; the two leading models pass about 63% of target tests but complete fewer than 8% of assignments end to end.
1 min · 240 words
Your Agent Will Pick the Hackable One
Geoff Stearns argues that as AI makes firmware cheap to write, owners will replace the software their devices ship with, AI agents will steer buyers toward hackable hardware, and makers like M-Vave opening up firmware show where consumer gadgets are headed.
6 min · 1,420 words
Agentic AI Designs an Analog Synthesizer
Tim tests whether frontier AI agents can 'vibe-engineer' real analog and mixed-signal hardware by having them design a touch-controlled synthesizer circuit and PCB, then building and evaluating the result.
8 min · 1,886 words
ESCAPES: a benchmark of 327 bugs that reached real users
breken's open ESCAPES benchmark rewinds 33 open-source products to 1 July 2026 code and scores 235 user-reported bugs: breken's run-based system caught 76 (32%) with 20 false alarms versus 37 for Claude Fable 5.1 in Claude Code; vendor-run results with public data and judge prompts.
16 min · 3,592 words
Deepnote's open benchmark drops 12 frontier models into the Slack of PageMantle, a synthetic company with 11 services, across 50 workplace situations. GPT-6 Astra leads at 8.97/10; models help well but struggle to stay silent, and pricier models aren't always better.
27 min · 6,246 words
Agent First Development for Enterprise Teams
An essay proposing "Agent First Development", a team methodology for shipping with coding agents: agree on requirements and interfaces up front, give agents context and freedom to execute, and keep humans responsible for every change that reaches customers.
9 min · 1,978 words
Running coding agents without handing them my credentials
Antonio describes lend, an open-source tool that lets remote coding agents use credentials, SSH keys and tools on his laptop only with approval, covering leases, audit logs, answering agent prompts remotely, what it does not protect against, and prior art.
10 min · 2,391 words
Advancing science & engineering with Antigravity's /teamwork
Google Research shares results from Antigravity's /teamwork multi-agent workflows, used with human experts to tackle long-open problems in mathematics, chemistry, systems software and ML engineering, and says the tools are being made available externally.
13 min · 3,076 words
Beating the frontier on NetHack
Antithesis explains how its exploration fuzzer, which can now find a path to victory within five wall-clock minutes on nearly 80% of NetHack seeds, got there, and how the same search-and-optimization approach applies to testing software.
77 min · 17,750 words
Gemini 4 Hydrogen and Orchestrator
Google Antigravity rolls out Gemini 4 Hydrogen plus a new Orchestrator model option that dynamically hands work between Gemini 4 Carbon (planning) and Hydrogen (execution) within one trajectory, reaching near-Carbon success on Terminal-Bench 4.1 with 24% less quota and 19% faster completion.
3 min · 633 words
The Brave Search API now accepts agentic payments via the x402 protocol
Brave announces that its Search API now supports pay-as-you-go agentic payments via the open x402 protocol: agents pay per request in USDC on Base (about $0.005 per query) with no account or API key, using a 402 Payment Required challenge and signed payment retry.
2 min · 419 words
Google introduces Gemini 4 Hydrogen, a fast, cost-efficient model in the Gemini 4 family built as a daily driver for everyday coding and reliable agentic workflows, with availability across the Gemini API, Gemini app, Search and Workspace.
8 min · 1,894 words
AWS Emulator and the engineer's job when agents write the code
Swami Sivasubramanian, in a guest post on Werner Vogels' blog, recalls simulating Amazon's Paxos lock service Alf and explains how AWS Emulator, a local copy of 200+ AWS services written and tested entirely by agents, puts infrastructure inside the coding loop.
7 min · 1,690 words
My AI Agents Found an Unmapped Maya Site
A first-person account of pointing a team of AI agents at NASA lidar data and, in under 72 hours, surfacing a previously unreported Maya site in Mexico with a plaza, pyramid and platforms, then getting an official reply from the Mexican government.
6 min · 1,390 words
An essay arguing that subagents spawned from a summary lose crucial context and are often slower and costlier than letting the main agent do the work; compares summary-based and forked subagents and describes an experimental 'smart fork' approach.
8 min · 1,759 words
The dawn of the Data Intelligence Engineer
MotherDuck's Jordan Tigani argues that AI agents now handle much of traditional data-team work, and proposes a new role, the Data Intelligence Engineer, who owns the data model, curates context, and makes sure the numbers agents produce are right.
10 min · 2,271 words