frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: Yansu – agent that builds your bespoke app before you ask

https://yansu.app/
2•yubozhao•1h ago
Hi HN

Most teams run on a dozen SaaS tools and use maybe 20% of each. The rest is someone else's workflow that you're paying for.

We built Yansu to fix that. It learns how you work — your desktop, your Slack, your Teams — and proactively builds bespoke tools tailored to you. You don’t need to prompt. It sees a pattern, builds a solution, and surfaces it for you to use or toss.

Like, I do the same things before every call: look through Linkedin, see any open threads on email, read the latest blog/news of the company and any relevant slack conversations. Yansu noticed and built me an app that pulls all of the information together. I can always browse it before / during / after the call now.

How Yansu works:

Listen — learns from your desktop activity and messaging apps Crystallize — distills patterns into structured knowledge about how you work Solve — builds tools based on that knowledge, without being asked

It builds apps/GUIs when you need to see things, CLIs when you need to automate or script things. Depends on what it learns about how you work.

Everything processes locally on your machine. When it needs an LLM, it sends structured context to whatever model you configure, be it Claude or GPT.

Free tier: bring your own agent, no credit card. $20/mo if you want managed AI. Available from today on macOS, Windows, Linux.

US national debt surges past $39 Trillion

https://apnews.com/article/us-national-deficit-hits-39-million-6ff73495bae701b5c009d3da5515ca3a
1•Betelbuddy•45s ago•0 comments

CliHub – The software user is shifting from human to agent. The store should too

https://github.com/clihub-ai/clihub
1•ziyangxie•56s ago•0 comments

Why AI Isn't HAL 9000

https://blog.protovate.ai/why-ai-isnt-hal-9000/
1•janadiamond•1m ago•0 comments

Ask HN: How do you work in teams with AI tools?

1•vjvjvjvjghv•2m ago•0 comments

US F-35 damaged by suspected Iranian fire makes emergency landing

https://www.cnn.com/2026/03/19/politics/f-35-damage-iran-war
1•mrtksn•2m ago•0 comments

Show HN: The Agent Skills Standard – A modular approach to LLM context

https://medium.com/@muhammad.shafat/stop-engineering-prompts-start-engineering-context-a-guide-to...
1•muhammad-shafat•2m ago•0 comments

We monitor internal coding agents for misalignment

https://openai.com/index/how-we-monitor-internal-coding-agents-misalignment
1•surprisetalk•4m ago•0 comments

BigQuery Graph Series Tutorial: Build a Graph from Unstructured Data

https://medium.com/google-cloud/bigquery-graph-series-6a768ccb351b
1•tanelpoder•5m ago•0 comments

Measuring LLM generation before token commitment

https://github.com/IvY-Rsearch/precomit
1•IvY-Rsearch•5m ago•0 comments

Show HN: A small app that ships with AI agent skills to extend and audit it

https://system-design-math-practice.tzvi.dev
1•tadwork•6m ago•0 comments

Show HN: Perstack – Containerized harness, 5 tests with full logs and API cost

https://github.com/perstack-ai/perstack
1•FL4TLiN3•8m ago•0 comments

Playing with dolls can help children develop imagination and social skills

https://www.theguardian.com/society/2026/mar/18/playing-with-dolls-can-help-children-develop-imag...
1•adrianhon•9m ago•1 comments

First of a Kind – Agentic CTF - with prizes!

https://jemini.live
1•4ppsec•10m ago•2 comments

Launch HN: Voltair (YC W26) – Drone and charging network for power utilities

3•wweissbluth•11m ago•0 comments

Show HN: A text-based space trading game that runs in Telegram

1•yaronr•11m ago•0 comments

Scaling Karpathy's Autoresearch: What Happens When the Agent Gets a GPU Cluster

https://blog.skypilot.co/scaling-autoresearch/
5•hopechong•12m ago•0 comments

Analysing the First Solid-State Battery [video]

https://www.youtube.com/watch?v=WNX47A8BysE
1•doctoboggan•12m ago•1 comments

New vibe coding experience – Google AI Studio

https://twitter.com/GoogleAIStudio/status/2034654985850659149
3•barbolo•12m ago•0 comments

CellState: React isn't the terminal UI bottleneck, the output pipeline is

1•nathan-cannon•15m ago•0 comments

Show HN: AI Digital Audio Workstation

https://sonurastudio.com/
1•kindred•16m ago•0 comments

Rawq – semantic code search for AI agents (4x fewer wasted tokens, Rust, OSS)

https://github.com/auyelbekov/rawq
1•Yerzhigit•16m ago•0 comments

Claude Code Experiment: Visualizing Hybrid Logical Clocks

http://muratbuffalo.blogspot.com/2026/03/claude-code-experiment-visualizing.html
5•mad44•18m ago•0 comments

Roundcube Snoozed Messages Support

https://packagist.org/packages/poremland/snoozed_messages
1•oremlandp•19m ago•0 comments

Three anesthesia drugs all have the same effect in the brain

https://news.mit.edu/2026/three-anesthesia-drugs-all-have-same-effect-brain-0317
2•gmays•20m ago•0 comments

Can Before and After Be Superposed?

https://physics.aps.org/articles/v19/38
1•bikenaga•20m ago•0 comments

A survey on LLMs for spreadsheet intelligence

https://orbilu.uni.lu/handle/10993/67962
4•teleforce•21m ago•0 comments

King opens world's longest coastal path around England

https://www.bbc.co.uk/news/articles/cy0dxexdd8xo
2•mellosouls•21m ago•0 comments

The age verification status of Open Source Operating Systems

https://github.com/BryanLunduke/DoesItAgeVerify
3•LorenDB•22m ago•1 comments

Versioned repo files as shared memory for multi-agent coding workflows

https://github.blog/ai-and-ml/github-copilot/how-squad-runs-coordinated-ai-agents-inside-your-rep...
2•spboyer•25m ago•0 comments

Communication Is Surveillance by Design

https://idiallo.com/blog/communication-is-surveillance-by-design
3•Brajeshwar•25m ago•0 comments