frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Show HN: Infinate –O(k)constant-time spatial attention for unlimited LLM context

https://github.com/ch1pu/infinate
1•ch1pu•50s ago•0 comments

ChatGPT needs a persistent workspace layer

2•LostBeacon•58s ago•0 comments

Show HN: PFAS Reporting 2026 – Triage tool for the EPA's 2011–2026 lookback rule

https://pfasreporting2026.com
1•todaycompanies•2m ago•0 comments

NexPhone: Android, Linux, and Windows 11 in a single device

https://nexphone.com/
1•gessha•2m ago•0 comments

Show HN: Find what you should play next

https://gamebrain.co
1•dsky•3m ago•0 comments

The Truth About Toilet Swirl

https://www.smartereveryday.com/toiletswirl
1•ColinWright•3m ago•0 comments

Get Closer So I Can Hear the Birds

https://terratauri.com/blog/get-closer-so-i-can-hear-the-birds/
2•terramauthe•4m ago•0 comments

Show HN: Ably AI Transport – a transport layer for agentic apps

https://ably.com/docs/ai-transport
1•mchristensen•6m ago•0 comments

Face Shape Detector

https://whatismyfaceshape.net/
1•adilblati3•7m ago•1 comments

Show HN: Tandem – open-source cross-platform AI coworker (Tauri)

1•frumu•11m ago•1 comments

Show HN: TerabyteDeals – Compare storage prices by $/TB

https://terabytedeals.com
1•vektor888•11m ago•0 comments

Oldest Cave Art Discovered in Indonesia

https://www.barrons.com/news/world-s-oldest-cave-art-discovered-in-indonesia-3ddd2a48
1•wjb3•11m ago•0 comments

Typeframe PS-85 – portable inspired by Alien

https://www.typeframe.net/docs/ps-85
2•vpribish•12m ago•1 comments

Show HN: An Ishihara (colorblindness) test generator

https://connorboyle.io/ishihara/
1•connorboyle•13m ago•0 comments

What if AI is both good and not that disruptive?

https://deadneurons.substack.com/p/what-if-ai-is-both-really-good-and
1•nr378•13m ago•0 comments

Grok exposes detailed account suppression information

https://twitter.com/realstewpeters/status/2013848860771144075
3•gravisultra•14m ago•1 comments

TV Screen Type Explained [video]

https://www.youtube.com/watch?v=8x1imoz5nrs
1•debesyla•15m ago•0 comments

Yes, code review is the bottleneck, no you're not doing it right

https://cognition.ai/blog/devin-review
1•agtestdvn•15m ago•1 comments

eBay Explicitly Bans AI "Buy for Me" Agents in User Agreement Update

https://www.valueaddedresource.net/ebay-bans-ai-agents-updates-arbitration-user-agreement-feb-2026/
3•bdcravens•17m ago•0 comments

My First Product Launch

https://veloblanco.global/
1•TBaronG•17m ago•1 comments

NeurIPS accepted research papers with 100 AI-hallucinated citations

https://fortune.com/2026/01/21/neurips-ai-conferences-research-papers-hallucinations/
1•Jimmc414•17m ago•0 comments

Spain's rail network under scrutiny after second deadly crash in three days

https://www.theguardian.com/world/2026/jan/21/spain-rail-network-scrutiny-after-second-train-crash
1•geox•19m ago•0 comments

I Built a Localhost Tunneling Tool in TypeScript – Here's What Surprised Me

https://softwareengineeringstandard.com/2026/01/20/i-built-a-localhost-tunnel-in-typescript-and-h...
1•aussieguy1234•23m ago•0 comments

Stevey's Birthday Blog

https://steve-yegge.medium.com/steveys-birthday-blog-34f437139cb5
2•squirrel•24m ago•0 comments

Beam Bots – Resilient Robotics

https://beambots.dev/
3•lawik•25m ago•0 comments

High-Availability Feature Flagging at Databricks

https://www.databricks.com/blog/high-availability-feature-flagging-databricks
3•bcongdon•26m ago•0 comments

Safety, Efficiency, and Workload of Predictive Display in Simulated Teledriving

https://www.mdpi.com/1424-8220/26/1/221
1•PaulHoule•30m ago•0 comments

Comparing 15 AI video models side-by-side using identical prompts

https://lmarena.ai/?chat-modality=video
2•leanvector•31m ago•2 comments

A macOS cache cleaner for browser and dev and AI caches (Clean / DeepClean)

https://clutterfall.app
1•darekhta•32m ago•2 comments

Spotify won court order against Anna's Archive, taking down .org domain

https://arstechnica.com/tech-policy/2026/01/annas-archive-said-spotify-scrape-didnt-cause-domain-...
7•voxadam•33m ago•2 comments
Open in hackernews

Show HN: CausaNova – Deterministic runtime for LLM constraints via Ontology

https://petzi2311.github.io/
1•CausaNova•1h ago

Comments

CausaNova•1h ago
Hello HN,

I watched Alex Karp’s recent talk at Davos/WEF where he stated: "Half your enterprise doesn't work on the battlefield. It works in a lab on a PowerPoint."

As a self-taught architect fighting German bureaucracy, I felt that. So I built an engine to fix the "Execution Gap" between LLM capabilities and strict operational constraints.

The Solution: CausaNova I built a deterministic .NET 8 runtime (Zero-Dependency) that acts as a firewall between the "Intent" (AI) and the "Action" (Reality).

It is already domain-agnostic. While the demo shows a kinetic defense scenario, the engine doesn't care about the target. We have already implemented ontologies for:

    HTML: Rendering strict UI components.

    SQL: Enforcing safe query states.

    Math: Solving geometric constraints.

    PLC/Hardware: controlling signals. To CausaNova, these are just domains in a JSON DSL.
How it works:

    Logic over Code: We stripped business logic out of the C# code entirely.

    Ontology as Config: Rules are defined in a strict JSON schema.

    The Resolver: The runtime uses an internal solver (SMT-style logic) to validate every request against the current state before execution.

    Tactical Edge: No cloud required. Runs on low SWaP hardware.
Why? Because "Ethics" is a debate, but "Logic" is a binary state. I wanted a system where I can hot-reload the "Rules of Engagement" (JSON) without recompiling the kernel.

The Stack: .NET 8 (AOT), Zero external dependencies.

I’m looking for feedback on the ontology structure and offline-inference strategies. Roast my architecture.