1-Bit Bonsai, the First Commercially Viable 1-Bit LLMs

TE

Telem

TL;DR: Web search open router that routes agent web search across providers (Exa, Parallel, Tavily, Brave, SerpAPI etc.), and traces web search results with quality metrics, so you can visualize whether a bad agent run is a web search problem or a reasoning problem. I work in venture capital, and before that I worked in information retrieval. When agents went viral, I built a due diligence agent for my day job. I vibe-coded the first version. It was pretty bad. My first reaction is to blame the model. I tried to change backend models, maybe because Fable doesn't want to help me do people search? Maybe because GPT 5.5 overthinks things? I tried DeepSeek, Kimi, Qwen; I changed prompts again and again, no good. Then I stopped just simply checking the final answer and started reading the actual trajectories: what did the agent search for? What came back? What pages did it read? At what point agents were off the rail? Pretty interesting: Sometimes the agent is stuck and confused because the information provided is irrelevant or even wrong; Sometimes the agent keeps iterate the same query, like "XXX lab UCB CS PhD founder 2026" "2026 XXX lab machine learning systems students startup" because the web search provider is not up-to-date; Sometimes sub-agents give up too early because the agent's intelligence is not enough or too "guardrailed"; Sometimes everything is just slow. A run jeopardized at minute 1 but still ran for another 10 before returning nonsense. That's because of bad searches, but nothing reveals it. So I built two things: 1. Web Search/Fetch router: One gateway for Exa, Parallel, Tavily, Brave, Ceramic, Linkup, Seltz, You, SerpAPI, etc. You or your agent can pick one or query several concurrently, and the responses come back in a homogeneous format. 2. Web Search Observability: Trace every search your agent (and its sub-agents) makes with an evaluator scoring relevance, diversity, and a few other things. Point is to answer one question: is your pipeline broken because the web search is bad? If so, where? Everything is agent-operated. All u need to do is to run this shell script: <curl -fsSL https://docs.telem.ai/alpha_install.sh | sh> Please give it a try, any feedback/suggestion would be great! Especially I’d like to ask: how are you debugging search/retrieval failures nowadays?

VG

Voronoi Go

I posted once before but wanted to share again because a lot has changed. There's now a fairly strong bot to play against (contributed by a community member) and also correspondence games. The combination of these things helps a bit in finding people to play against.

WB

We built the smallest dual-band aircraft tracker

We've been building open source embedded ADS-B receivers for a while, and spent the past 8 months smallifying our existing receiver tech with a new chip from Semtech. Ask me anything about ADSB or hardware manufacturing!

RI

Risklytics

Hi HN, we're Sam and Alex, founders of Risklytics (https://risklytics.ai). We're both on leave from Harvard, and we run an insurance brokerage for companies building robots, drones, autonomous systems and satellites. You describe your machine, we find the insurance companies willing to cover it, and we read everything before you sign. Buying insurance for any frontier tech goes like this: - your first customer or pilot requires coverage before your machine can work on their site - a broker hands you an application written for ordinary businesses, and often will deny you on baseless "risks" (one of our clients got denied for using CAD software in their workflow) - you get filed under something close or outright denied, the person at the insurance company often can't tell what they're looking at, and the safe answer is no - if a quote does come back, the AI risk is often quietly removed by an add-on clause that prevents you from truly being covered Those clauses surprised us. Insurance companies mostly don't write their own policies from scratch. They license standard forms from an industry body called ISO and attach add-on clauses to adjust them, and in January ISO published clauses that void coverage for losses arising out of AI in any part of a process. Each insurer decides whether to adopt them, account by account, and there's no public record of who has and to what capacity. There's an open question in the wording too. It defines generative AI as systems trained on data that produce text, images, audio, video or code. Whether a robot's control model, which produces motor commands, falls under that is unsettled. We've asked underwriters, the people at insurance companies who make these calls, and gotten different answers across the board. We didn't initially plan to be brokers. We came into YC to work in insurance and spent the first half of the batch on a wildfire risk model we wanted to sell to insurance companies. They liked it and wanted to build their own, which is the polite way to say no. While that was dying, a batchmate who launches satellites mentioned he couldn't get insured, and once we started asking around we kept hearing the same story from hardware companies. So we got licensed in a week and now can place coverage. What we've built so far: - a six-step application that asks about what you're building and drafts the structured version insurance companies like, omitting anything you didn't say - a map of which insurers add the AI clauses, which will remove them, and which cover frontier companies on purpose - when an insurer says no, you see their reason and what would change it, full transparency all the time - a hands off process for companies trying to get fully covered Here's a demo of the whole process: https://youtu.be/gVqCamTNH5E On money: the insurance company pays us a commission when you get covered, usually 10-20% of what you are charged, and we charge you nothing on top (that's right, no brokerage fees). Also, if you're a software company with ordinary needs, we'll still handle the whole process for you and get you the best policy we can find. In the past week alone we have helped three companies get covered that were denied elsewhere. The companies include a robotics wholesaler and rental store, who had trouble getting covered earlier because some of their property uses AI (even though they never use it themselves), a bridge collision system that was getting denied because they use CAD software, and a tele-operating robotics service that needed coverage within a week for their first pilot. If you're looking for insurance in any capacity feel free to reach out to me at alex@risklytics.ai or fill out an application at https://risklytics.ai. And if you work on the carrier side and think you'd wanna partner with us let us know.

HM

How much of Hacker News is about AI?

This website counts titles containing the standalone word "AI", case-sensitive and word-bounded. "OpenAI" doesn't count. "AI-powered" does. There's a toggle for a wider vocabulary: artificial intelligence spelled out, LLM, GPT, and model and vendor names. Some quick stats: - 14.4% of new HN titles this year, against 10.9% in 2025. On the wider vocabulary, 21.9% and 16.1%. - First month above 1% was Oct 2016. Above 5%, Feb 2023. Above 10%, May 2025. - Over the last five years the quietest month was Jan 2022 at 0.8%. The peak was Feb 2026 at 15.9%. - 13.6% on weekdays against 12.4% at weekends. - Show HN runs highest of any category at 19.8% this year, which makes this post part of the number. As of this post, today is at 15% for the standard filter and 19% for the extended.