Nobie
Show HN: Nobie – an Excel-compatible runtime for agents and humans
I built a one-prompt hackathon platform, free entry, sponsored prizes
Show HN: I built a one-prompt hackathon platform, free entry, sponsored prizes
Microphone
If you are an aspiring founder, any VC will ask you this question: “why are you the only person who could solve this”. If you want to generate passive income with your side idea, get ready to enter a crowded market as everyone and their mother is shipping. Unless you have an active X account or you’re a TikTok sensation distribution is going to be tough. I just launched the trie.dev microphone beta to help folks find their edge. You yap into your phone about your ideas; Trie turns the rambling into hypotheses, then prioritizes them based on your experience and your realistic ability to distribute in that idea space — surfacing the problems only you can solve. From there you can generate creative and run Meta ads against your hypotheses straight from your phone, with zero setup, to see how real people respond. I built it initially for myself and friends as an “intake form” for running paid ads to help validate our side gig ideas. Happy to chat about how it works or the stack. Joining the waitlist will send you an email to join via TestFlight.
Benchmark your eng team's AI agent maturity in 5 minutes
we had hundreds of discussions with engineering leaders over the past few months, and everyone's trying to understand where they are in the AI journey. we collected all this data into a benchmark and built a free grader to let you know where you stand. you answer on a 1–5 scale (e.g., autonomy runs from "suggestions only" to "agents own multi-hour workflows across code, infra, and external systems") - takes about 5 minutes. https://agent-benchmarks.com/software-factory/ waiting for your results!
ZenStack
Hi HN, I'm Jiasheng, co-creator of ZenStack. Access control usually ends up scattered across app code instead of living with the data model — and that's riskier when the code is written by an agent, since it's easy to ship a query with a missing or incomplete authorization check. ZenStack enforces policy (RBAC/ABAC/relation-based) directly at the ORM layer, on top of Kysely, so every query gets checked the same way regardless of who wrote it and who is calling it. Postgres RLS is an option too, but it's hard to maintain and scale, and it's Postgres-only — ZenStack's approach is database-agnostic. One of our users, MermaidChart, put it well after launching their team feature on ZenStack: "much cleaner and easier to maintain than writing RLS policies or application-level checks that will surely leak after some time." Happy to answer anything.
Opening lines of famous literary works
This came from an idea that had been knocking around in my head for several years. I had been collecting opening lines of famous works and thought it would be cool to see one everyday as I opened the browser. I tried different styles but landed on the simple background with the text, let the words speak for themselves. Over time i've added more quotes I believe now there are close to 60, so hopefully you can refresh a few times and get a fresh one every time. I hope you guys like it, enjoy!
Oodle.ai
Hi HN, we're Kiran and Vijay! Over the past two years, we have built a columnar storage engine for observability: logs, metrics, and traces. Today, it's exciting for us to show what we've built on top of that foundation: LLM Agent Observability. Given how non-deterministic agents are, storing all traces without sampling was critical for us. But these traces tend to be in the MBs, sometimes GBs - we needed to store them inexpensively. We also needed the queries and analyses to be fast. To meet both these goals, we store them in S3 in our own parquet-like file format, and query them using AWS Lambda. Since we process each span of every trace, instead of running LLM-based evals on each, we first analyze them using deterministic techniques. We detect tool failures, retries, loops, abnormal token usage, latency regressions, schema violations, sentiment, and other production signals. We've written more about the approach here: https://blog.oodle.ai/you-cant-sample-your-way-to-reliable-a... The combination of our own engine, no sampling, and deterministic processing before LLM-for-evals allows us to price at $10 per million traces, provide sub-second p99 query latency, and have healthy margins. Before building this, we used Langfuse for our own agent observability, which was 6x more expensive. Still super early, and rough around some edges, we would love your questions and feedback!
Real-time avatars that change emotions as you talk
Hey HN, we're Ben and Caoimhe, cofounders of Anam. We build interactive avatars and just shipped our latest model, cara-4. This is the first avatar model which can naturally shift emotions and expressions during the conversation; a feature we’re calling “Director Notes”. The way it works is relatively simple: we have an LLM provide cues such as [laughter], [sad], [warm] interleaved with the speech, which we then condition our animation model with. To test it out, we commissioned a blind study from Mabyduck.com with 200 participants, 1,600 rated live interactions across six criteria. Cara-4 ranked first overall and was preferred head-to-head over each competitor on overall experience, lip-sync, visual quality and “naturalness”. On latency, measured across a full week of live traffic, end of user-speech to first video frame is ~1.2s median. The avatar model's own share is just ~100ms; most of the rest is waiting on STT, LLM, TTS or various forms of buffering (an unsung latency killer). How the model works: cara-4 has a two-stage design, a diffusion transformer turns audio+text into motion embeddings (head pose, gaze, lip shape, expression), and a rendering model applies those to a reference image, so new faces works without finetuning. Why faces at all: they carry emotional signal that text and voice don't, and they're a more accessible medium. Anam started in part from Ben watching his gran struggle with her iPad and thinking there should be a face she could just talk to. If you’d like to test it for free go to anam.ai
Agnost
Hey HN, we’re Shubham & Parth, childhood friends building Agnost AI (https://agnost.ai), product analytics for teams building chat and voice agents. We read production conversations and find behavioral failures like users rageprompting (cursing at the agent), repeatedly rephrasing the same request, correcting the agent, asking for missing features, or leaving after an answer that was technically successful. We have an interactive demo with no signup here: https://app.agnost.ai?demo=true Here's a demo video: https://www.tella.tv/video/agnost-ai-launch-hn-demo-9haa The core problem is that chat and voice products do not have the same metrics as web apps. When the product interface is language, clicks and funnels become much less useful. Users also rarely give explicit feedback, and when they do it's usually sugarcoated. I barely type /feedback in Claude or Codex myself. Most users just curse, ask again, correct the agent, or leave. So product engineers get technical visibility from latency, errors, and traces, but still have to guess whether users got what they wanted. We got here after building around agents for the last year and got a couple of founders asking for something like a PostHog for conversations for the AI assistants they were building. We are not trying to be in the observability or evals space. Observability tells you what happened technically. Evals validate cases you already know. We're more on the discovery side like what users wanted, where they got frustrated, what they asked for repeatedly, and what new evals should exist. Teams send us agent conversation messages through SDKs or OTel, optionally with metadata like account, plan, source, organization, etc. We cluster conversations into product-specific intents. Feature requests and bugs are default categories; most other clusters are created dynamically from the customer’s data and evolve over time. You can create your own cluster in plain English. If a cluster gets too broad, we split it. If a new pattern appears, we suggest it. One AI video editor company used Agnost AI to find feature requests hidden inside chat. The biggest one was that around 70 users wanted auto-subtitles, but users said it as “add this text in this frame” 12x in a single session, “can you caption it”, “give me transcript of audio” and variations across languages. The team later built the feature. Doing this over millions of messages without sending everything to an LLM was the hard part initially. In ClickHouse, “fetch the last 50 events by time across conversations” and “fetch all events in this conversation” want different sort orders, so we had to iterate a lot on sorting keys, partitions, materialized views, and projections. For finding new clusters, sending everything through an LLM was too slow and expensive. HDBSCAN-style embedding clustering also gets painful at scale because of pairwise comparisons. We first split conversations into segments based on cosine drift, run BIRCH to compress the candidate space, and then use HDBSCAN-like clustering on the smaller set. For matching existing clusters, we use embeddings, smaller classifiers/BERT-style models, and LLMs only as fallback for ambiguous cases. We’re live with multiple companies and ingesting ~1M chat and voice messages per day. Pricing is public: Starter is free, Pro is $499/month, and Enterprise is for higher volume, security, retention needs. We use each customer’s data only for that customer. We are SOC 2 Type 1 compliant, Type 2 is in progress, and our SDKs are on PyPI and npm. We’d love feedback from the HN community and people building chat or voice agents: how do you detect these signals today, what feedback methods have worked, and what would block you from trying this? Happy to answer questions and take criticism.
A Free RSS reader with a configurable recommendation engine
Show HN: A Free RSS reader with a configurable recommendation engine
Leet Robotics: Learn robotics and ROS2 with hands-on courses
Hi all, I've just launched Leet Robotics: a platform to learn robotics hands-on, with a full ROS2 workspace that runs in the browser (Jazzy, Gazebo Harmonic, Foxglove, VS Code) - no install required. The platform also has room for sharing projects and simulation assets as it grows. Our first course is live now: Intro to ROS2 (free to read). The course teaches skills ranging from building your first node to a capstone project of a robot touring a museum world, with every lesson runnable in the online workspace (free accounts get an hour of workspace time daily - enough to follow the course). Would love feedback from this community: on the course, the workspace experience, and what courses to build next.
Aproov
Aproov is a marketplace that matches early-stage founders with creators for distribution deals Match with the RIGHT creators for your company, offer whatever you want/have (premium access, cash, revenue share, co-created content, other), gain real distribution, repeat. Aproov verifies each step of the process We created this because with our previous company, distribution was a real pain. So we decided to help other founders so they don't spend as much effort as we did, and they could get better and faster results and focus purely on building