Fovea
Video in, your JSON out — validated against your schema
What it does
Fovea turns a video into JSON that matches a schema you supply. Send a video and a JSON Schema — or a rough sketch of one — to one endpoint, or add the MCP server to Claude Code or Codex in a single line. Every response is validated against your schema before it is returned; a malformed shape is retried internally, never handed back half-right. Any field can carry a timestamp. Videos are deleted once the job finishes. 30 free minutes, no card.
Fovea turns a video into JSON that matches a schema you supply. Send a video and a JSON Schema — or a sketch of one — to one endpoint, or add the MCP server to Claude Code or Codex. Every response is validated against your schema before it is returned. 30 free minutes, no card.
Upload a video. Hand us a JSON schema. Get back exactly the structured data your workflow expects. Nothing more, nothing to configure. Muybridge needed a row of tripwired cameras to turn a gallop into sixteen numbered frames. Frames are cheap now. Turning them into something your code can read is the part still worth paying for. Fovea is a video-to-JSON API. You send a video and the shape you want back — a JSON Schema, or a sketch of one — and you get JSON that conforms to it, validated before it leaves us. It comes three ways: a web app, one POST to /v1/extract , and an MCP server that Claude Code, Codex or any MCP client installs in a line. Any field in your schema can carry a timestamp.…from fovea.run
Does the same job
all alternatives →- PJPg_jsonschema – A Postgres extension for JSON validation2022 · github.com · ▲203
pg_jsonschema is a solution we're exploring to allow enforcing more structure on json and jsonb typed postgres columns. We initially wrote the extension as an excuse to play with pgx, the rust framework for writing postgres extensions. That let us lean on existing rust libs for validation (jsonschema), so the extension's implementation is only 10 lines of code :) https://github.com/supabase/pg_jsonschema/blob/fb7ab09bf6050... happy to answer any questions!
- VJVerify JSON using minimal schema2020 · github.com · ▲91
- JSJSON storage bins with schema validation2018 · npoint.io · ▲74
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, August 2026
the whole month →- TL
Life & fun · 10d ago · louisabraham.github.io
- SA
Hello HN! I found that picking out plausible but diverse skin tones for my digital art and game development projects was kind of difficult, and I got curious about if there was a way to define a color space that made it easy. I've built a color picker and procedural generation algorithm based on the space as well as a bunch of other fun js features and demos throughout the page that use the equations. If you find it interesting, I have lots of explanations of how I built it and what properties the space has. The methodology might be a bit shaky, but hopefully the result is as helpful for…
Life & fun · Aug 2026 · toneyalexander.github.io
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 16d ago · simedw.com