frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Bevy 0.19

https://bevy.org/news/bevy-0-19/
1•paavohtl•3m ago•0 comments

Before SpaceX IPO, Investors in China Acquired Stakes

https://www.propublica.org/article/spacex-elon-musk-ipo-foreign-investors-china
1•ilreb•3m ago•0 comments

I Stored a Website in a Favicon

https://www.timwehrle.de/blog/i-stored-a-website-in-a-favicon/
1•theanonymousone•9m ago•1 comments

Gizmodo breached, hosting malware and no action taken for hours

https://old.reddit.com/r/cybersecurity_help/comments/1uajrup/beware_gizmodo_captcha_malware_attempt/
1•mullingitover•17m ago•1 comments

Virtual whiteboard for sketching hand-drawn like diagrams

https://github.com/excalidraw/excalidraw
1•modinfo•23m ago•0 comments

Bun 1.4 coming July 7th

https://twitter.com/jarredsumner/status/2068059134952677887
1•sorenbs•26m ago•0 comments

Show HN: I wrote a Rust book ending with a Redis clone

https://shankhan3.gumroad.com/l/dnwmtp
1•zeeshanali0094•28m ago•0 comments

Ask HN: What's a simple app you'd build if you had a weekend?

2•akashwadhwani35•29m ago•0 comments

GenAIDojo

https://www.genaidojo.io/
1•aniketwattawmar•42m ago•0 comments

EAS Alarm at midnight caused by a hacker in Brazil

https://www.reddit.com/r/mildlyinfuriating/s/wLRv2s1u5k
2•davidkwast•44m ago•2 comments

Study: Conservatives are dying at higher rates than liberals

https://www.fastcompany.com/91561329/widening-health-gap-between-liberals-and-conservatives
1•MilnerRoute•47m ago•0 comments

The New Calculus of AI-Based Coding

https://blog.joemag.dev/2025/10/the-new-calculus-of-ai-based...
2•aabdi•52m ago•1 comments

Designing a backyard deck for my house

https://blog.cosmin.cloud/posts/diy-deck.html
1•spycraft•57m ago•0 comments

Compress tool outputs, logs, files, RAG chunks before LLM for 60-95% less tokens

https://github.com/chopratejas/headroom
1•gmays•57m ago•0 comments

Trapped in a Waymo: SF passenger recounts terrifying construction zone ordeal

https://abc7news.com/post/waymo-recall-san-francisco-man-recounts-terrifying-moment-when-robotaxi...
1•mikhael•59m ago•0 comments

Show HN: Stocks Porfolio in a URL and Favicon

https://stocksreader.com/
2•franciscop•1h ago•1 comments

Bureaulogy – formation, evolution, and perpetuation of bureaucracies

https://grantslatton.com/bureaulogy
1•thelastgallon•1h ago•1 comments

Chromium Embedded Framework (CEF)

https://github.com/chromiumembedded/cef
1•auraham•1h ago•0 comments

The Punctum and the Blind Field

https://theholbrookreport.com/reports/punctum-and-the-blind-field/
1•TyrunDemeg101•1h ago•0 comments

China will have a Fable 5-class AI model before next year

https://www.tomshardware.com/tech-industry/artificial-intelligence/elon-musk-says-that-china-will...
5•achow•1h ago•0 comments

TV Player for Rocket Launches

https://github.com/sighmon/SpaceX-TV
1•sighmon•1h ago•0 comments

We record because we have forgotten how to remember

https://pilgrimsage.substack.com/p/the-lowest-frequency
1•momentmaker•1h ago•0 comments

AI and the Great CMS Unbundling

https://dri.es/ai-and-the-great-cms-unbundling
2•christefano•1h ago•0 comments

This pole is worse than any Flock Camera [video][38 mins]

https://www.youtube.com/watch?v=wEfbhEVuvMM
1•Bender•1h ago•0 comments

Lawsuit against Amazon over suicides linked to chemical can go to trial

https://www.cbsnews.com/news/amazon-suicide-case-sodium-nitrite-washington-supreme-court/
2•hentrep•1h ago•0 comments

Satellite reveals immense scale of GPS signal tampering

https://www.space.com/space-exploration/satellites/its-quite-a-bit-more-than-we-expected-satellit...
11•y1n0•1h ago•0 comments

Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance

https://hustvl.github.io/Moebius/
2•ilreb•1h ago•0 comments

Spirit Crossing's AI Problem

https://blog.curiousquail.com/spirit-crossings-ai-problem/
1•zokiboy•1h ago•0 comments

A Practical Guide to SSH Tunnels: Local and Remote Port Forwarding

https://labs.iximiuz.com/tutorials/ssh-tunnels
2•signa11•1h ago•0 comments

How to lose a war in three easy steps

https://www.ft.com/content/0ba6b511-649e-4a29-bb99-f824584ec93d
12•Alien1Being•1h ago•1 comments
Open in hackernews

Forked CozoDB to give agents cognitive primitives

1•shanrizvi•2h ago
CozoDB is an embedded relational-graph-vector database created by Ziyang Hu and the Cozo Project authors. It does Datalog, vector search (HNSW), and full-text search in a single embedded engine and was created with the vision to be the “Hippocampus for AI”. Unfortunately, the project went quiet after December 2024. KuzuDB, another embedded graph database with vector and full-text indexes similar to Cozo, was archived in October 2025.

I decided to continue the vision by using Cozo to build a memory and reasoning substrate with a cognitive ontology inspired by neuroscience. It covers events, epistemic structure, intent, and action, with multi-agent support. An extraction pipeline using an LLM of your choice to extract these from documents and session transcripts, deduplicates them, and detects cross-document contradictions. Everything is versioned and provenanced back to source chunks, with time travel.

But the features we rely on most needed work, so I hard-forked Cozo as MnesticDB under the same MPL-2.0 license, and made improvements.

I reduced index build time from 89s to 8s on a 40,000 chunk corpus with real 384-dimension embeddings without affecting recall by rebuilding the HNSW index flat-and-parallel rather than row-by-row.

I unblocked reads during index rebuilds, which would take more than an hour for a 151,000 index with 1536 dimensions; it now builds off-lock, with a 40,000 vector index building under 6 seconds with 90,000 concurrent reads in one test.

I added reciprocal-rank fusion across semantic (HNSW), keyword (BM25), and graph proximity signals at around 42ms, 4x faster than the three queries running separately.

After switching the default full-text scorer from tf-idf to Okapi BM25, fused recall improved from 0.75 to 0.954 on a 40,000 chunk corpus, on par with DuckDB’s 0.957.

MnesticDB: https://www.mnesticdb.com https://www.github.com/shuruheel/mnestic

Returning to the memory and reasoning substrate, I added support for structured data, allowing users to connect a Postgres database, draft an operational ontology from the schema, and sync rows in as typed graph nodes, with provenance and on-demand live fields, fused into the same graph extracted as the cognitive ontology. This allows an agent to query business tables and knowledge distilled from documents through one interface, with contradictions and history on both.

MindGraph: https://mindgraph.cloud/ https://www.github.com/shuruheel/mindgraph-rs

In a 60-question stratified pilot of LongMemEval, we scored 87% with zero retrieval errors, strong factual recall (100%), and strong temporal reasoning (90%). The misses clustered in preference questions (70%), which I subsequently fixed, but haven’t re-run the benchmark yet. The plan is to eventually run the full 500-question run as the confidence interval is quite wide at the current sample size.

I’ve been building this solo, which is not optimal, so if this piques your interest, please consider becoming a contributor to the MnesticDB project, or a co-founder of MindGraph (the cognitive graph substrate for agents).

Ask me anything here, or email shan@rizvi.nu.