think on Hacker News
91 titles from 87 authors in the last 30 days, with 78 distinct links and 6 Show HN.
Counts titles, it does not confirm facts. Open the source before concluding anything.
Projects on this term
The score comes from our model.
- Show HN: Make your AI agent think before it codes · score 6.8
SpecOps adds a structured 4-phase spec-driven workflow (understand, spec, implement, complete) to AI coding assistants via slash commands, with git-tracked markdown specs, multi-platform support, drift detection, and production learnings capture. It targets teams using AI coding agents who want to enforce requirements and design before implementation, and its novel production feedback loop and adversarial evaluation differentiate it from simpler prompt templates.
- Show HN: Local wiki Claude Code/Codex write for you to think, organize, direct · score 6.8
Nimbalyst Wiki is a Claude Code/Codex plugin that turns coding sessions into a local markdown wiki stored in your repository, capturing decisions, open questions, and tracked items. It offers a browser-based editor supporting drawings, mind maps, spreadsheets, and mockups, with multi-agent compatibility via MCP. It fills a gap in AI-assisted development by providing persistent, organized project memory that agents can write to and humans can review.
- Show HN: The Steam Frame's wireless adapter isn't what you think · score 6.3
A Rust tool that allows a PC to connect to the Steam Frame's hidden WiFi hotspot without Valve's official adapter by reverse-engineering Steam's Remote Play LAN protocol. It discovers the headset, sends custom SSID/password credentials, and connects via NetworkManager, enabling Multilink streaming with any compatible WiFi card. The project is a clever piece of reverse engineering with a detailed write-up, but its utility is niche due to hardware and setup requirements.
- Show HN: We simulated 9,297 Texans to explore what undecided voters think · score 6.0
This project simulates a population of 9,297 Texans using a combination of real polling, census data, and LLM-generated personas to explore the reasoning of undecided voters in a Senate race. It is interesting because it attempts to address known biases in LLM-based political simulation by grounding personas in quantitative models and real data, offering a novel interactive tool for understanding voter sentiment beyond simple polls.
Recent titles
- Scientists think our recent ancestors may have run hotter – but why?
2026-11-01 · bbc.com · HN discussion - Will AI take your job? You have more agency than you think
2026-10-31 · vox.com · HN discussion - Which hobby do you think is most diverting and pleasurable?
2026-10-31 · HN discussion - Neuralink's Newest Brain Chip Isn't What You Think [video]
2026-10-31 · youtube.com · HN discussion - Alkaline batteries are obsolete – but not for the reason you'd think [video]
2026-10-31 · youtube.com · HN discussion - I don't think AGI is right around the corner (2025)
2026-10-31 · dwarkesh.com · HN discussion - Free to Think
2026-10-30 · scholarsatrisk.org · HN discussion - Agentic Animation Tool – think After Effects for your agents
2026-10-30 · github.com · HN discussion - Show HN: We simulated 9,297 Texans to explore what undecided voters think
2026-10-30 · texas-midterms.theorea.ai · HN discussion - Sterling Anderson doesn't think humanoid robots should be building cars
2026-10-30 · theverge.com · HN discussion - Dutch startup Elysian says electric flight may be closer than you think
2026-10-29 · electrek.co · HN discussion - Qwen 3.8 invents URLs with real key ID, that's why I think it's no backdoor
2026-10-29 · xda-developers.com · HN discussion - Ask HN: How do you keep your mind from forgetting how to think?
2026-10-28 · HN discussion - What It Costs to Think
2026-10-28 · chrbutler.com · HN discussion - Reading and writing are less symmetric than you (probably) think
2026-10-28 · fgiesen.wordpress.com · HN discussion - Ask HN: Do u think there should be a auto recovery tool for agents?
2026-10-27 · HN discussion - Show HN: Make your AI agent think before it codes
2026-10-27 · github.com · HN discussion - Things to Think About Post-Election (US)
2026-10-27 · movementvoterproject.substack.com · HN discussion - Why it is foolish to think AI can't write like humans
2026-10-26 · HN discussion - Why AI-generated text is harder to understand than you think
2026-10-26 · unite.ai · HN discussion - Sam Altman: 'I No Longer Think We're Gonna Be Able to Just Patch All the Bugs'
2026-10-24 · gizmodo.com · HN discussion - "I no longer think we're gonna be able to just patch all the bugs" – Sam Altman
2026-10-24 · twitter.com · HN discussion - Attention people who think UB is fine
2026-10-24 · blog.habets.se · HN discussion - AI models think humans are a bunch of amoral sheep
2026-10-23 · theregister.com · HN discussion - The Proponents of 'Dark Nutrition' Think Everything You Know About Food Is Wrong
2026-10-23 · nytimes.com · HN discussion - UUIDv7 doesn't have the properties you think it does
2026-10-23 · alexsci.com · HN discussion - Local AI Is Not as Private as You Think [video]
2026-10-23 · youtube.com · HN discussion - Ask HN: What do you think about lightweight default games on Linux distros?
2026-10-23 · HN discussion - Why even the smartest people on the planet can think dumb things
2026-10-23 · fastcompany.com · HN discussion - Geoffrey Hinton: 'I think there's a 1% chance it will destroy us'
2026-10-22 · english.elpais.com · HN discussion