AI Engineering
Practical ideas and tools for building with AI.
About this list
This reading list brings together items from the sources selected for AI Engineering. Open each item at its original source for the complete article, episode or video. This is a selection of current coverage, not a complete archive.
Latest stories
Newest first
Llm-gemini 0.35
Release: llm-gemini 0.35 I added support to llm-gemini for the new Gemini 4 Argon model, made available today via the Gemini API and for customers of their $100/month subscription plan. Here are the low, medium and high pelicans. As expected, they are all excellent: Gemini has a track record of producing great SVGs, and are the only lab where I know their staff have invested in SVG specifically (and who have boasted about their pelicans before.) Here's the "high" illustration, which cost me 17.166 cents: The pricing is interesting.
Release rust/v1.0.22-preview.0 · github/copilot-sdk
Update SDK snapshot for Copilot CLI 1.0.98-0
Release langgraph-cli==0.4.34 · langchain-ai/langgraph
Changes since cli==0.4.33 release(cli): 0.4.34 (#9300) feat(cli): add --json to deploy listeners list (#9243)
Release v0.91.11 · github/gh-aw
🌟 Release Highlights This release focuses on reliability: sturdier safe outputs, better diagnostics, and fixes for Codex and Aider engines. ✨ What's New Claude Haiku 5.5 pricing is now supported (#68424). Work-queue issues now show work titles and descriptions (#68421), with unified Git ref resolution (#68425). push_to_pull_request_branch accepts numeric targets and gives clearer errors (
Bench 0.8.1-rc.2
Second pre-release of the Rust bench CLI, built from bench-rs at 6c3fa1d. Without BENCH_VERSION, the install script installs only a bench release that is not a pre-release. There is none yet, so it stops and asks you to set BENCH_VERSION. What changed since 0.8.1-rc.1 The cloud metadata endpoints are refused under the default open network and under a blocklist. What is refused: on Docker, Podman and Daytona, an agent that runs as a non-root user is refused TCP to 169.254.169.254 and the other cloud metadata addresses, on every port but 53.
Release v1.0.21 · github/copilot-sdk
Other changes improvement: [All SDKs] regenerate RPC and session event types against Copilot CLI 1.0.97 (new catalog agent plugin/media type fields and updated session event schemas) improvement: [Rust] JSON-RPC and session filesystem dispatch handling refinements improvement: [C#] minor update to built-in PermissionHandlers Full Changelog: v1.0.19...v1.0.21 Warning Firewall blocked 1 domain The following domain was blocked by the firewall during workflow execution: github.com
Release v1.0.21-preview.3 · github/copilot-sdk
Internal dependency updates only: this prerelease refreshes the SDK snapshot for Copilot CLI 1.0.97-3. There are no other user-visible changes since v1.0.21-preview.2. Warning Firewall blocked 1 domain The following domain was blocked by the firewall during workflow execution: github.com To allow these domains, add them to the network.allowed list in your workflow frontmatter: See Network Configuration for more information. Generated by
AI Engineering Playbook 🔥
Download My Playbook for Free
Building resilient AI-powered software, and planning for when things go wrong
How to handle uncertainty with AI vendors and disaster recovery (DR) approaches to consider. An excerpt from the book Building Resilient Distributed Systems by Sam Newman
Inside Design Studio: Giving Your Terminal a Tasteful Visual Director
In Part 1, I shared the painful realization many technical builders eventually stumbles into: you can’t prompt-engineer your way to good visual design. Natural language in a chat box is fundamentally
V1.6.4 — Vector storage & DB reliability fixes
v1.6.4 — Vector storage & DB reliability fixes Release Date: 2026-10-13 Changes: v1.6.3 → v1.6.4 Summary This patch release fixes several backend issues around vector search adapters and the Turso database adapter, and removes an unwanted background pruning task for LanceDB-backed stores. The changes reduce memory use and errors during searches, make Turso open failures report consistently, and add tests and harness fixes to improve reliability.
Release v4.1.24 · cline/cline
Changed With no model chosen, the Cline and OpenRouter providers now use Claude Sonnet 5.5. For Cline, that’s the first model in Cline’s recommended list. Selecting ai& shows its own API key and m
Release v0.11.0 · agentevals-dev/agentevals
What’s Changed feat: v1alpha1 protobuf contract + native gRPC by @krisztianfekete in #251 Give rubric_based_* metrics a public rubric path (#132) by @erauner12 in #229 fix: resolve the package ver
Otel-gpu-collector: 1.0.0
OpenTelemetry GPU Collector 1.0.0 Features Skips write disk metrics for read-only Linux block devices; reporting resumes when devices become writable. Missing or invalid read-only indicators are treated as writable. (#1397)
Release v1.0.19-unstable.38345923974.g3754b70 · github/copilot-sdk
Update SDK snapshot for Copilot CLI 1.0.95-0.unstable.r38345923974.g3
Quoting Geoffrey Huntley
If the agent isn't using my diagnostics library, I could introspect and update AGENTS.md, but there's still a delta, and it'll drift. Instead, I write a pre-commit hook that checks the git diff for direct calls to observability modules that bypass my diagnostics module. Boom: fail the commit, inject a prompt telling the agent to use the diagnostics library. All of a sudden it's ten out of ten.
V1.6.4.dev0 — Vector search reliability & Turso error handling
v1.6.4.dev0 — Vector search reliability & Turso error handling Release Date: 2026-10-13 Changes: v1.6.3 → v1.6.4.dev0 Pull Requests: #5745 Summary This small development release fixes a key vector-search bug and improves Turso database error handling. It also updates a few packaging dependencies and tests so search is more reliable and errors are handled consistently. Highlights Fixed a vector-search bug that prevented selecting raw vector data during search (pgvector and Lancedb adapters) — reduces bandwidth and avoids errors.
Why could the same AI rebuild a 16,000-line program without seeing its code, yet fail to customize a library for METR? The answer shapes what you hand it
Watch now (21 mins) | Are we developing a more general intelligence, or getting much better at producing extraordinary abilities that transfer unevenly?
Why software engineering for deterministic computers is the wrong control plane for agents
Announce Cohort 2 of Harness Engineering Class with Packt
Release gen/go/v0.11.0 · agentevals-dev/agentevals
chore(ui): bump dev dependencies to clear npm audit findings (#255)
Showing 20 of 190 articles
Know a source we’re missing? Tell us about it