Real problems people complain about online, pulled every morning and scored out of 100. Build, validate, or skip. How scoring works
Hey HN, we're Willy and Dan, co-founders of Twill.ai (https://twill.ai/). Twill runs coding CLIs like Claude Code and Codex in isolated cloud sandboxes. You hand it work through Slack, GitHub, Linear, our web app or CLI, and it comes back with a PR, a review, a diagnosis, or a follow-up question. It loops you in when it needs your input, so you stay in control.Demo: https://www.youtube.com/watch?v=oyfTMXVECbsBefore Twill, building with Claude Code locally, we kept hitting three walls1. Parallelization: two tasks that both touch your Docker config or the same infra files are painful to run locally at once, and manual port rebinding and separate build contexts don't scale past a couple of tasks.2. Persistence: close your laptop and the agent stops. We wanted to kick off a batch of tasks before bed and wake up to PRs.3. Trust: giving an autonomous agent full access to your local filesystem and processes is a leap, and a sandbox per task felt safer to run unattended.All three pointed to the same answer: move the agents to the cloud, give each task its own isolated environment.So we built what we wanted. The first version was pure delegation: describe a task, get back a PR. Then multiplayer, so the whole team can talk to the same agent, each in their own thread. Then memory, so "use the existing logger in lib/log.ts, never console.log" becomes a standing instruction on every future task. Then automation: crons for recurring work, event triggers for things like broken CI.This space is crowded. AI labs ship their own coding products (Claude Code, Codex), local IDEs wrap models in your editor, and a wave of startups build custom cloud agents on bespoke harnesses. We take the following path: reuse the lab-native CLIs in cloud sandboxes. Labs will keep pouring RL into their own harnesses, so they only get better over time. That way, no vendor lock-in, and you can pick a different CLI per task or combine them.When you giv
Hacker News5mo agoToolAI
I saw the HackerRank (YC S11) hiring post (https://news.ycombinator.com/item?id=47667011) and it made me realize I no longer understand how to evaluate candidates effectively.Specifically, we are changing hiring across 3 dimensions: > Tasks: Real-world tasks on code repositories vs standard algorithmic-style puzzles > Evaluation: AI fluency, orchestration skills vs functional correctness > Candidate experience: Agentic IDE vs a simple code editorIn the “old world,” you could ask multiple questions and triangulate skill from answers. Now it seems like evaluation depends heavily on tools and models that keep changing month to month.So I’m curious: > What signals actually correlate with strong engineers today? > How do you design interviews that don’t become obsolete with the next model release? > Are algorithmic interviews still useful at all?Would love to hear from people who have recently changed their hiring process or have been interviewed using this new approach.
Hacker News5mo agoToolAI
Hi HN! I've just released CSS Studio, a design tool that lives on your site, runs on your browser, sends updates to your existing AI agent, which edits any codebase. You can actually play around with the latest version directly on the site.Technically, the way this works is you view your site in dev mode and start editing it. In your agent, you can run /studio which then polls (or uses Claude Channels) an MCP server. Changes are streamed as JSON via the MCP, along with some viewport and URL information, and the skill has some instructions on how best to implement them.It contains a lot of the tools you'd expect from a visual editing tool, like text editing, styles and an animation timeline editor.
Hacker News5mo agoToolAI
Hi HN! We're excited to share marimo pair [1] [2], a toolkit that drops AI agents into a running marimo notebook [3] session. This lets agents use marimo as working memory and a reactive Python runtime, while also making it easy for humans and agents to collaborate on computational research and data work.GitHub repo: https://github.com/marimo-team/marimo-pairDemo: https://www.youtube.com/watch?v=6uaqtchDnocmarimo pair is implemented as an agent skill. Connect your agent of choice to a running notebook with:/marimo-pair pair with me on my_notebook.pyThe agent can do anything a human can do with marimo and more. For example, it can obtain feedback by running code in an ephemeral scratchpad (inspect variables, run code against the program state, read outputs). If it wants to persist state, the agent can add cells, delete them, and install packages (marimo records these actions in the associated notebook, which is just a Python file). The agent can even manipulate marimo's user interface — for fun, try asking your agent to greet you from within a pair session.The agent effects all actions by running Python code in the marimo kernel. Under the hood, the marimo pair skill explains how to discover and create marimo sessions, and how to control them using a semi-private interface we call code mode.Code mode lets models treat marimo as a REPL that extends their context windows, similar to recursive language models (RLMs). But unlike traditional REPLs, the marimo "REPL" incrementally builds a reproducible Python program, because marimo notebooks are dataflow graphs with well-defined execution semantics. As it uses code mode, the agent is kept on track by marimo's guardrails, which include the elimination of hidden state: run a cell and dependent cells are run automatically, delete a cell and its variables are scrubbed from memory.By giving models full control over a stateful reactive programming environment, rat
Hacker News5mo agoToolAI
Keeper is an embeddable secret store (Argon2id, XChaCha20-Poly1305 by default). Four security levels, audit chains, crash-safe rotation. Vault is overkill for most use cases. This is for when you ge paranoid about env and need encrypted local storage that doesn't suck. No security through obscurity, hence, It's still early, so now's the best time to find weird edge cases, race conditions, memory leaks, crypto misuse, anything that breaks. The README has a full security model breakdown if you want to get adversarial.
Hacker News5mo agoToolAI
Businesses using vibecoded (AI-built) custom tools face ongoing issues with feature requests, bugs, and API changes, leading to frustration without developer support.
X8mo agoOtherAI
AI users with surplus token quotas at certain times can't lend or borrow them to others facing shortages, leading to wasted capacity.
X5mo agoToolAI
AI platforms like Pixverse are frustrating for creators as they alter custom characters (e.g., changing user's elf design) in generated videos, ruining workflows.
X5mo agoToolAI
When teams use AI agents today, they operate in silos—running in separate tabs, terminals, or applications—making it impossible for humans to see what agents are doing, coordinate across multiple agents, or maintain control over their outputs. This creates coordination chaos, duplicated work, and loss of oversight. Teams either avoid using agents altogether or struggle through manual context-switching and fragmented workflows.
Product Hunt5mo agoToolAI
Hey HN! We are Bharath, and Simranjit from Relvy AI (https://www.relvy.ai). Relvy automates on-call runbooks for software engineering teams. It is an AI agent equipped with tools that can analyze telemetry data and code at scale, helping teams debug and resolve production issues in minutes. Here’s a video: [[[https://www.youtube.com/watch?v=BXr4_XlWXc0]]]A lot of teams are using AI in some form to reduce their on-call burden. You may be pasting logs into Cursor, or using Claude Code with Datadog’s MCP server to help debug. What we’ve seen is that autonomous root cause analysis is a hard problem for AI. This shows up in benchmarks - Claude Opus 4.6 is currently at 36% accuracy on the OpenRCA dataset, in contrast to coding tasks.There are three main reasons for this: (1) Telemetry data volume can drown the model in noise; (2) Data interpretation / reasoning is enterprise context dependent; (3) On-call is a time-constrained, high-stakes problem, with little room for AI to explore during investigation time. Errors that send the user down the wrong path are not easily forgiven.At Relvy, we are tackling these problems by building specialized tools for telemetry data analysis. Our tools can detect anomalies and identify problem slices from dense time series data, do log pattern search, and reason about span trees, all without overwhelming the agent context.Anchoring the agent around runbooks leads to less agentic exploration and more deterministic steps that reflect the most useful steps that an experienced engineer would take. That results in faster analysis, and less cognitive load on engineers to review and understand what the AI did.How it works: Relvy is installed on a local machine via docker-compose (or via helm charts, or sign up on our cloud), connect your stack (observability and code), create your first runbook and have Relvy investigate a recent alert.Each investigation is presented as a notebook in our web UI, with data visualizat
Hacker News5mo agoToolAI
I love C and C++, but setting up projects can sometimes be a pain.Every time I wanted to start something new I'd spend the first hour writing CMakeLists.txt, figuring out find_package, copying boilerplate from my last project, and googling why my library isn't linking. By the time the project was actually set up I'd lost all momentum.So, I built Craft - a lightweight build and workflow tool for C and C++. Instead of writing CMake, your project configuration goes in a simple craft.toml: [project] name = "my_app" version = "0.1.0" language = "c" c_standard = 99 [build] type = "executable" Run craft build and Craft generates the CMakeLists.txt automatically and builds your project. Want to add dependencies? That's just a simple command: craft add --git https://github.com/raysan5/raylib --links raylib craft add --path ../my_library craft add sfml Craft will clone the dependency, regenerate the CMake, and rebuild your project for you.Other Craft features: craft init - adopt an existing C/C++ project into Craft or initialize an empty directory. craft template - save any project structure as a template to be initialized later. craft gen - generate header and source files with starter boilerplate code. craft upgrade - keeps itself up to date.CMakeLists.extra.cmake for anything that Craft does not yet handle.Cross platform - macOS, Linux, Windows.It is still early (I just got it to v1.0.0) but I am excited to be able to share it and keep improving it.Would love feedback. Please also feel free to make pull requests if you want to help with development!
Hacker News5mo agoToolAI
AI agents struggle to interact with desktop apps like humans due to OS not designed for AI control; need purpose-built OS.
X6mo agoToolAI
Small business owners and marketers struggle to create authentic website copy that resonates with their target customers. Traditional website builders offer templates with generic placeholder text that requires extensive rewriting and still fails to capture what genuinely differentiates their business. They lack the time, copywriting skills, or customer insight to articulate their unique value proposition, resulting in websites that sound like every competitor and fail to convert visitors into customers.
Product Hunt5mo agoToolAI
Hacker News5mo agoToolAI
I've been frustrated one too many times by terrible airplane wifi and not being able to load Gmail or Superhuman when all I want to do is get a few simple text-only emails out the door.These clients have become pretty bloated with the assumption you've always got great bandwidth.So I vibe coded BAREMAIL. It's open source, has no backend, and you can just set it up for yourself. Takes ~3 mins to setup API access via Google Cloud Platform (thanks for making this not super easy Google!)I tried to maintain nice design and some important keyboard shortcuts without getting to overBEARing.
Hacker News5mo agoToolAI
I've spent the last years obsessed with a sideproject to build a humanistic 3D modeler for desktop.By humanistic I mean a tool that stays out of your way, instead of requiring the user to learn both a complex UX surface as well as a complex theoretical basis before being able to model effectively. The GUI is uncrowded and the modeling affordances are only those which are intuitive to present to the user. Which is sort of backwards compared to most CAD packages where the technical complexity takes the front stage. Here the hierarchy is intentionally reversed.This is still in alpha-stage, but the features are mature enough for feedback and experimentation.TinkerCAD is actually what comes philosophically closest to this, but it's hobbled either by Autodesk's strategy or by technical limitation to be a really good tool beyond certain complexity.The ambition here eventually is to provide a tool that has same intuitive capability as building Lego bricks, while not compromising on engineering qualities.The main intent is to make extrusion based modeling operations super easy, to offer robust STL and STEP import and allow complex modeling via boolean operations.The modeling logic is parametric and volume based - the surface presentation is always a discretized water tight triangle mesh.This is the clearest philosophical differentiator to traditional CAD/CAM packages - or visual editors like Blender. Rather than force the user to nurse surface topology at every stage, the modeler will only permit those operations that result in a correct output.This is not an SDF (signed distance field) modeler. The domain model is fully based on parametric analytic shapes. This means the tessellation is crisp and specific.The modeling data is immutable and serialized to disk while modeling. For the user this gives a perfect undo and zero data loss.It's built for efficiency first - my test workhorse is a Thinkpad T14 Gen 2 i5 with an integrated gpu.It's not su
Hacker News5mo agoToolAI
About six months ago, I started working on a project to fine-tune Whisper locally on my M2 Ultra Mac Studio with a limited compute budget. I got into it. The problem I had at the time was I had 15,000 hours of audio data in Google Cloud Storage, and there was no way I could fit all the audio onto my local machine, so I built a system to stream data from my GCS to my machine during training.Gemma 3n came out, so I added that. Kinda went nuts, tbh.Then I put it on the shelf.When Gemma 4 came out a few days ago, I dusted it off, cleaned it up, broke out the Gemma part from the Whisper fine-tuning and added support for Gemma 4.I'm presenting it for you here today to play with, fork and improve upon.One thing I have learned so far: It's very easy to OOM when you fine-tune on longer sequences! My local Mac Studio has 64GB RAM, so I run out of memory constantly.Anywho, given how much interest there is in Gemma 4, and frankly, the fact that you can't really do audio fine-tuning with MLX, that's really the reason this exists (in addition to my personal interest). I would have preferred to use MLX and not have had to make this, but here we are. Welcome to my little side quest.And so I made this. I hope you have as much fun using it as I had fun making it.-Matt
Hacker News5mo agoToolAI
I built this because I wanted to see how far I could get with a voice-to-text app that used 100% local models so no data left my computer. I've been using a ton for coding and emails. Experimenting with using it as a voice interface for my other agents too. 100% open-source MIT license, would love feedback, PRs, and ideas on where to take it.
Hacker News5mo agoToolAI
ProblemHunt5mo agoToolAI
We’re Ben and Jacob, cofounders of Freestyle (https://freestyle.sh). We’re building a cloud for Coding Agents.For the first generation of agents it looked like workflows with minimal tools. 2 years ago we published a package to let AI work in SQL, at that time GPT-4 could write simple scripts. Soon after the first AI App Builders started using AI to make whole websites; we supported that with a serverless deploy system.But the current generation is going much further, instead of minimal tools and basic serverless apps AI can utilize the full power of a computer (“sandbox”). We’re building sandboxes that are interchangeable with EC2s from your agents perspective, with bonus features:1. We’ve figured out how to fork a sandbox horizontally without more than a 400ms pause in it. That's not forking the filesystem, we mean forking the whole memory of it. If you’re half way down a browser page with animations running, they’ll be in the same place in all the forks. If you’re running a minecraft server every block and player will be in the same place on the forks. If you’re running a local environment and an error comes up in process that error will be there in all the forks. This works for snapshotting as well, you can save your place and come back weeks later.2. Our sandboxes start in ~500ms.Demo: https://www.loom.com/share/8b3d294d515442f296aecde1f42f5524Compared with other sandboxes, our goal is to be the most powerful. We support full Linux + hardware-virtualization, eBPF, Fuse, etc. We run full Debian with multiple users and we use a systemd init instead of runc. Whatever your AI expects to work on debian should work on these vms, and if it doesn’t send a bug report.In order to make this possible, we’ve moved to our own bare metal racks. Early in our testing we realized that moving VMs across cloud nodes would not have acceptable performance properties. We asked Google Cloud and AWS for a quote on their bare metal nodes and found tha
Hacker News5mo agoToolAI