frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: Dewey – Ingest docs, search semantically, get cited AI answers

https://meetdewey.com
2•lambdabaa•1h ago
Flat chunking throws away document structure. A PDF isn’t a bag of paragraphs. It has sections, subsections, and a hierarchy that carries meaning. An agent that can’t navigate that structure can’t do serious research.

I ran into this building RAG over scientific literature. The standard approach (embed chunks, find top-k, generate) works fine for simple Q&A but falls apart when you need real research depth: multi-hop reasoning across papers, synthesizing conflicting results, tracing a finding back to the exact passage in a methods section. The problem wasn’t the models.

Dewey treats documents, sections, and chunks as first-class API primitives. The section manifest (full heading hierarchy with titles and byte offsets) lets agents scan cheaply before committing to full chunk retrieval, the same way a researcher skims a table of contents before reading. The /research endpoint runs an agentic loop; at “exhaustive” depth it can traverse an entire corpus, iteratively query, and return a grounded answer with numbered inline citations pointing to the exact source passage.

Two ways in:

- REST API + TypeScript/Python SDKs for developers building research or document Q&A into their apps - MCP server (@meetdewey/mcp on npm) for anyone using Claude, ChatGPT, or Cursor. Your document collections become tools without writing any code.

Bring your own OpenAI key and depth becomes a quality setting rather than a billing one. That includes AI image captioning, which makes figures and diagrams searchable alongside your text. No markup on generation.

Built this solo. Happy to answer questions about the architecture, the retrieval design, or anything else. Curious whether others have found section-aware retrieval makes a meaningful difference vs. flat chunking in practice.

Free tier, no credit card required: https://meetdewey.com

Law Firms Prefer Cubicles to Cubicle Dwellers

https://b2bs.substack.com/p/law-firms-prefer-cubicles-to-cubicle
1•oopsiremembered•2m ago•0 comments

Encoding Team Standards

https://martinfowler.com/articles/reduce-friction-ai/encoding-team-standards.html
1•saikatsg•2m ago•0 comments

Show HN: ModelAtlas – Find AI models that HuggingFace search can't

https://github.com/rohanvinaik/ModelAtlas
1•Vinaik•2m ago•1 comments

Noyb win: Microsoft ordered to stop tracking school children

https://noyb.eu/en/noyb-win-microsoft-ordered-stop-tracking-school-children
1•jruohonen•3m ago•0 comments

Kevin Rose Back at Digg

https://www.kevinrose.com/p/rebooting-everything
1•minkeymaniac•4m ago•0 comments

Jami – free/libre, end-to-end encrypted, and private communication software

https://jami.net/
1•smartmic•5m ago•0 comments

Show HN: I adapted codex-plugin-cc's design for Gemini CLI's ACP

1•abiswas97•5m ago•0 comments

Major Claude Code source leak offers deep insight into how Anthropic tool works

https://arstechnica.com/ai/2026/03/entire-claude-code-cli-source-code-leaks-thanks-to-exposed-map...
1•johnbarron•8m ago•0 comments

Why Inventing Color TV Was So Difficult [video]

https://www.youtube.com/watch?v=hyjCmIbRRvs
1•DamnInteresting•9m ago•0 comments

After 16 years and $8B, military new GPS software still doesn't work

https://arstechnica.com/space/2026/03/after-16-years-and-8-billion-the-militarys-new-gps-software...
2•johnbarron•10m ago•0 comments

Employers Beware: Uptick in BIPA Lawsuits Targeting AI Note-Taking Software

https://www.amundsendavislaw.com/labor-employment-law-update/employers-beware-uptick-in-bipa-laws...
3•jruohonen•11m ago•0 comments

Textstring

https://pushmatrix.github.io/textstring/
1•memalign•11m ago•0 comments

Show HN: Wozcode – double Claude Code output

https://www.wozcode.com/
4•bcollins34•15m ago•0 comments

Show HN: Spotlytt, A platform to create video resumes

1•pbs29•16m ago•2 comments

Jailbroken, open-source, pre-built iOS 26 virtual machine

https://github.com/34306/vphone-aio
2•huragok•16m ago•0 comments

A survey of how companies are buying AI professional services

https://www.aienablementinsider.com/products/buyer-survey
1•dylancollins•17m ago•0 comments

SwarmSync– We built the transaction layer for autonomous AI Commerce

https://swarmsync.ai
1•TaxFix•18m ago•1 comments

Visual Pipeline Builder

https://www.pipevals.com/
1•todsacerdoti•20m ago•0 comments

Claude Code Buddy Creator

https://carter.works/experiments/claude-code-buddy-creator/
1•open-paren•20m ago•2 comments

Google attributes Axios hack to North Korea

https://techcrunch.com/2026/03/31/hacker-hijacks-axios-open-source-project-used-by-millions-to-pu...
6•ariporad•20m ago•0 comments

Ask HN: I burnt out from software development. What now?

2•fnoef•21m ago•2 comments

GitHub's Historic Uptime

https://damrnelson.github.io/github-historical-uptime/
10•todsacerdoti•21m ago•2 comments

VinFast going all-in on electric scooters with battery swap rollout in Vietnam

https://electrek.co/2026/03/31/vinfast-is-going-all-in-on-electric-scooters-with-massive-battery-...
1•breve•21m ago•0 comments

What about juniors?

https://brooker.co.za/blog/2026/03/25/ic-junior.html
1•lr0•23m ago•0 comments

LLMs encourages delusional thinking in patients, study finds

https://www.theguardian.com/technology/2026/mar/14/ai-chatbots-psychosis
3•lr0•25m ago•0 comments

Two more Liquid Glass fixes in macOS 26.4

https://anderegg.ca/2026/03/30/two-more-liquid-glass-fixes-in-macos-264
1•speckx•25m ago•0 comments

Do LLMs Break the Sapir-Whorf Hypothesis?

https://dnhkng.github.io/posts/sapir-whorf/
1•dot_treo•27m ago•3 comments

When an ALB Can Replace Nginx (and When It Can't)

https://yaw.sh/blog/when-alb-replaces-nginx/
2•jeffyaw•27m ago•0 comments

Ask HN: Is there any founder building non AI startup in 2026?

1•daudmalik06•27m ago•0 comments

GitHub Will Train Copilot Models with User Data by Default

https://www.heise.de/en/news/Only-with-opt-out-GitHub-will-train-Copilot-models-with-user-data-in...
2•lschueller•28m ago•1 comments