Hackernews posts about TCAS

1. Claude Fable 5: mid-tier results on coding tasks (www.endorlabs.com)
   410 points by bugvader 25 days ago | 250 comments
2. You can't unit test for taste (dev.karltryggvason.com)
   306 points by kalli 13 days ago | 140 comments
3. Hey, n00b, we didn't hire you to complete tasks (newsletter.kentbeck.com)
   229 points by rrvsh 17 days ago | 128 comments
4. Build a Basic AI Agent from Scratch: Long Task Planning (medium.com)
   142 points by ruxudev 27 days ago | 53 comments
5. Show HN: I built 184 free browser tools – PDF, image, dev, AI tasks, no upload (brevio.pro)
   89 points by ruimbarreira 20 days ago | 32 comments
6. Show HN: Lightweight Task queue on Erlang/OTP, SQLite-backed, no overengineering (github.com)
   75 points by ent1c3d 27 days ago | 18 comments
7. MiniMax M3 vs. GLM 5.2: Codegen comparison across autonomous coding tasks (thinkwright.ai)
   55 points by oceanwaves 17 days ago | 19 comments
8. Task Failed Successfully: Saturating NIC and Disk Bandwidth (blog.mrcroxx.com)
   52 points by MrCroxx 13 days ago | 20 comments
9. Fable 5 will default to Opus 4.8 for coding tasks (xcancel.com)
   49 points by babelfish 5 days ago | 30 comments
10. Claude Fable 5 will sabotage "frontier LLM research" tasks (twitter.com)
    49 points by qwertyforce 27 days ago | 6 comments
11. GLM-5.2: Built for Long-Horizon Tasks (z.ai)
    32 points by meetpateltech 20 days ago | 2 comments
12. NanoShell: A preemptively multi-tasked 32-bit OS with a windowed GUI (github.com)
    17 points by mrunix 5 days ago | 1 comments
13. Observation is the layer under taste. We built it for AI runtime (twitter.com)
    14 points by Cherepukhin 18 days ago | discuss
14. Show HN: Paper UX for task management, that organizes your personal DB itself (notaru.ai)
    11 points by alpadurza 20 days ago | 1 comments
15. Stolen French fries are spicier than justice: How covert larceny enhances taste (www.sciencedirect.com)
    10 points by bookofjoe 20 days ago | 8 comments
16. Why LLMs still lack taste (beyondtheprior.com)
    10 points by supermdguy 27 days ago | 5 comments
17. Show HN: Agentic coding workflows built on Git worktrees and task evidence (github.com)
    10 points by alexreysa 16 days ago | 2 comments
18. GLM-5.2: Built for Long-Horizon Tasks (huggingface.co)
    10 points by victormustar 20 days ago | discuss
19. The Story of PHP. Documentary Teaser [video] (www.youtube.com)
    9 points by pjmlp 22 days ago | 1 comments
20. Show HN: Macro – unified system for email, chat, tasks, docs, agents (AGPL/Rust) (github.com)
    8 points by jbecke 21 days ago | 1 comments
21. Show HN: TaskPeace – a task queue my AI coding agents pull work from over MCP (taskpeace.com)
    7 points by JulianQuinn 3 days ago | 7 comments
22. Throwing My Roku in the Trash (www.coryd.dev)
    7 points by cdrnsf 13 days ago | 2 comments
23. MiMo Code: Scaling coding agents to long-horizon tasks (mimo.xiaomi.com)
    6 points by tvvocold 26 days ago | 2 comments
24. Heinz tomato ketchup and the sweet taste of market dominance (www.economist.com)
    6 points by bookofjoe about 15 hours ago | 1 comments
25. Does MSG Make Food Taste Better? [video] (www.youtube.com)
    6 points by Cider9986 23 days ago | 1 comments
26. Taste turns your Git history into enforceable coding judgment (github.com)
    6 points by dvcoolarun 16 days ago | discuss
27. Hey, N00B, We Didn't Hire You to Complete Tasks (newsletter.kentbeck.com)
    6 points by RebootStr 18 days ago | discuss
28. Stolen fries are spicier than justice: How covert larceny enhances taste (www.sciencedirect.com)
    6 points by gnabgib 21 days ago | discuss
29. Ask HN: Does taste matter when writting code
    5 points by rafaepta 19 days ago | 3 comments
30. AI doesn't take jobs. It takes tasks (www.nextgig.rocks)
    5 points by poseidon_sg 11 days ago | 2 comments