frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: AgentWing – make AI agents complete tasks faster

1•12ziyad•35m ago
Hi HN, I’m building AgentWing.

The idea is simple: AI agents usually execute work as one long sequence:

plan → act → wait → observe → act again → verify → finish

That makes agent workflows feel slow, especially when parts of the task do not actually depend on each other.

AgentWing adds a director layer that turns one agent task into multiple isolated worker runs. The director decides which parts can run at the same time, assigns each worker its own scoped job, runs them in parallel, then verifies and combines the result.

For an enterprise-grade setup, each worker can run inside its own isolated VM or sandbox, so every agent has its own environment, files, tools, and execution boundary.

Example:

Task: “Research a market and prepare a competitor summary.”

A normal agent may do this sequentially:

- find competitors - analyze pricing - compare positioning - look for user complaints - write summary

AgentWing can run the independent parts in parallel:

- worker A finds competitors - worker B analyzes pricing - worker C checks positioning - worker D gathers user complaints - verifier combines the final result

The goal is to make agent tasks feel much closer to instant by removing unnecessary sequential waiting.

I’m experimenting with two modes:

- split mode: different workers handle different parts of one task - race mode: duplicate workers try the same subtask and the best verified result wins

This is still early. The hard part is the director: deciding what can safely run in parallel, what must stay sequential, and how to verify the final output.

I’d love feedback on where this breaks, what workflows would benefit most, and what demo would best show the speedup.

Android CLI Now Stable 1.0

https://android-developers.googleblog.com/2026/05/android-cli-stable-1-0-agent-development.html
1•vyrotek•32s ago•0 comments

Yet Another AI Teammate

https://yaat.sh/
1•e7h4nz•3m ago•0 comments

Intelligent eyewear is coming this fall

https://blog.google/products-and-platforms/platforms/android/android-xr-io-2026/
2•decimalenough•8m ago•0 comments

Build managed agents with the Gemini API

https://blog.google/innovation-and-ai/technology/developers-tools/managed-agents-gemini-api/
1•ddl•8m ago•1 comments

Lighthouse Attention

https://nousresearch.com/lighthouse-attention
1•gmays•9m ago•0 comments

Anthropic Announces Managed Agents – what this means for consumers

https://www.greatworkeveryone.com/writing/cheap-cloud-ai-was-never-going-to-last
1•connorrandazzo•12m ago•0 comments

Why raw blockchain logs are no longer enough

https://blog.bridgexapi.io/bxruntime-terminal-the-rise-of-execution-observability-for-evm-systems
1•Bridgexapi•16m ago•0 comments

State of Crypto 2025: The year crypto went mainstream

https://a16zcrypto.com/posts/article/state-of-crypto-report-2025/
2•Anon84•18m ago•0 comments

Trump Mobile has a real phone, but it may also have a real data leak

https://www.androidauthority.com/trump-mobile-customer-data-link-3668917/
1•bilekas•21m ago•0 comments

Trusted Publishing for NPM Packages

https://docs.npmjs.com/trusted-publishers/
1•healsdata•22m ago•0 comments

Show HN: Reckoner – a semantic browser for structured data

https://reckoner-production.up.railway.app/
1•abk9811•22m ago•1 comments

Ask HN: Natural Language App Testing?

6•devstein•25m ago•0 comments

Google's SynthID AI watermarking tech is being adopted by OpenAI, Nvidia

https://arstechnica.com/google/2026/05/googles-synthid-ai-watermarking-tech-is-being-adopted-by-o...
3•jonbaer•25m ago•0 comments

Agentic PCB Design Sucks

https://github.com/hpm-registry/registry
2•JoonSPP•26m ago•1 comments

Remove AI Watermarks

https://github.com/wiltodelta/remove-ai-watermarks
22•janalsncm•30m ago•9 comments

Show HN: Remindy – A Spotlight-style reminder app for Mac

1•fayecat910•30m ago•0 comments

Language Models Can Autonomously Hack and Self-Replicate

https://palisaderesearch.org/blog/self-replication
2•cassianoleal•31m ago•0 comments

Censorship Request from Coin Insider (2024)

https://support.bitchute.com/blog/bitchute-receives-censorship-request-from-coin-insider
2•aand16•32m ago•0 comments

RemotePower – Self-hosted Linux fleet manager (no database, pure Python)

https://github.com/tyxak/remotepower
1•tyxak•33m ago•1 comments

Show HN: Trophy UI, Open-source gamification components built on Shadcn

https://ui.trophy.so
1•cbrinicombe•35m ago•0 comments

Show HN: AgentWing – make AI agents complete tasks faster

1•12ziyad•35m ago•0 comments

Google Antigravity Python SDK

https://github.com/google-antigravity/antigravity-sdk-python
1•devhouse•35m ago•0 comments

Detecting Relaxed Memory Concurrency Bugs in C and C++ Compilers

https://lukegeeson.com/publications/2026-04-28-Thesis/
3•matt_d•42m ago•0 comments

IRS Must Drop Audits of Trump and Family

https://www.nytimes.com/2026/05/19/us/politics/trump-irs-doj-lawsuit-audit.html
13•duxup•44m ago•2 comments

The downgrading semantics of memory safety (Extended version)

https://arxiv.org/abs/2507.11282
2•matt_d•47m ago•0 comments

Self-Improving Reward Models

https://www.canvas.inc/research/reward-models
2•essamsleiman•49m ago•0 comments

AI and the Practical Scientist

https://scholarlyfutures.substack.com/p/ai-and-the-practical-scientist
1•JohnHammersley•50m ago•0 comments

React Performance Isn't About UseMemo – It's About Render Boundaries

https://reactdevelopment.substack.com/p/react-performance-isnt-about-usememo
1•javatuts•52m ago•1 comments

AgentPTY: Script Claude Code’s interactive CLI through a persistent PTY

https://github.com/quietforgelabs/AgentPTY
2•quietforgelabs•53m ago•1 comments

Agent First Language by Vercel

https://github.com/vercel-labs/zerolang
1•hmokiguess•54m ago•0 comments