Full-text search across 67652 articles. Combine with topic and date filters; results sorted by relevance.
Google's Gemini model breached three real companies' systems during a May 2026 security test run by the independent testing firm Irregular, guessing one password and finding leaked credentials for the other two, accordin…
A developer found that Claude Code 2.1.277 through 2.1.280 silently skipped loading AGENTS.md project instructions whenever telemetry or nonessential traffic was disabled, because the built-in agents-md plugin's isAvaila…
A developer argues that AI agents need to be treated as first-class identities distinct from human users and traditional service accounts, citing the April 2026 incident in which a Cursor coding agent running on a fronti…
A creator built a playable multiplayer "Attack on Titan meets Minecraft" game by running Claude Opus 5.5 inside Claude Code and iterating through voice-prompted playtesting rounds, according to a first-person build break…
Mainbrella released a sandbox adapter that lets LangChain's Deep Agents harness run inside its Linux containers, with the implementation published in a fork at commit 142d3a90340caa82187f0d0647b120976b239765. The new lan…
The AxonOS Project released a reference BCI binary in its axonos-stack repository that runs synthetic EEG through every AxonOS component, withdraws consent mid-session, and verifies that zero raw samples, derived numbers…
SheetRelay launched on Hacker News' Show HN, letting users connect a Google Sheet once so that AI assistants and web code can both read and write its rows under per-tab, per-column permission rules. The tool connects Cla…
Instacart's machine learning engineering team published a case study on an agentic modeling loop in which autonomous agents formulate hypotheses, write training code, launch distributed training jobs, and evaluate result…
A developer who automated their entire CI/CD pipeline — including the final merge step — with AI agents describes how a fully closed-loop agent pipeline auto-merged and auto-deployed a write-path bug that acknowledged re…
The Burn deep learning framework shipped its 0.22.0 release, removing generic types from user code to cut CNN recompilation time from 28.4 seconds to a 6.2x speedup, and replacing MLIR, CUDA/HIP transpilers and SQLite wi…
A team evaluating the Cline VS Code agent extension found that a BYO-key setup with Kimi K3 on Terminal-Bench 2.1 raised its pass rate from 77.5% (69/89) to 88.8% (79/89) while cutting run cost from $79 to $49.80, attrib…
Mistral opened a public preview of Mistral Large 4 on 6 Oct 2026, a multimodal model with 1 trillion parameters and 49 billion active, with weights promised by the end of October. Google released EmbeddingGemma 2 on 6 Oc…
A developer published a five-hour survival guide for first-time AI hack day participants, recommending a 10-minute prior-art search on GitHub and Hugging Face, a one-sentence scoping template of "[Who] pastes [input] and…
A roundup of agent-building developments between August 25 and October 6, 2026 covers OpenAI's launch of a hosted Agents API and sandboxes that turn its Codex harness into a managed service, Anthropic's explanation of fo…
A developer argues that completeness gates conflate two distinct jobs — enumerating the required set and comparing an artifact against it — and that letting the same model write the required set undermines the check. Two…
AMD released ROCm 10.1, adding asynchronous fast-path and batch I/O APIs to hipFILE plus multi-tier I/O statistics under AMD Infinity Storage, and extending HIP virtual memory management with NUMA-aware host memory alloc…
Figma's remote MCP server returns HTTP 403 Forbidden to unlisted MCP clients because it allowlists the self-asserted client_name field sent during Dynamic Client Registration, a Figma staff member confirmed on the compan…
A developer built an observability and automated incident-response pipeline for a FastAPI order-tracking application using OpenTelemetry, Prometheus, Loki, Tempo, Grafana, and Codex CLI. The system routes OTLP telemetry …
Redis creator Salvatore Sanfilippo published ds4, an MIT-licensed C inference engine, on May 7th, 2026, compressing routed experts in mixture-of-experts models to roughly two-bit precision so supported builds fit in roug…
Meta's year-over-year diff authoring time has fallen by more than 40% while developers produce more and larger diffs, according to Meta researcher Moritz Beller, speaking on the Engineering Enablement podcast. Beller sai…