We can't post as fast as we're shipping π’
Live this week:
- Hosted MCP server
- Claude Code plugin
- Codex plugin
- ChatGPT connector
- pip install relevan
- $20 Hobby plan
Search over your data, built for agents.
https://t.co/Nzisx0TQbP
87 PRs opened, 44 merged, 130k lines of code. Last 24 hours.
Speed is dangerous, so we tuned our workflow to make sure everything we ship:
- matches its original intent
- is secure
- won't take down prod π
How are you shipping faster?
https://t.co/ZDDHgUnse8
This is so true from my experience. Writing well is something we're still tuning Relevan for.
We have a few skills that get us close but we're a long way away from Claude writing everything.
@RelevanDev got accepted into AWS Accelerate π
We're building search that learns: every index tuned on its own data, every change tested before it ships, all on AWS.
Next up: π’π’π’ All the innovation
Thanks @OpenSearchProj and @AWS for the support!
Introducing Firecrawl MCP v3.
Give your agents access to all of the knowledge they need.
β 3000+ tools with automatic provider waterfalls
β 60x higher rate limits
β 4x more usage
β Spawn subtasks for parallel work
Available now for every harness.
https://t.co/3c4LhoiJ4y
Doing experiments is like the backbone of the search industry. Relevan is just doing them several times a day.
One step closer to a autonomous relevance engine.
We tested 5 relevance tricks on 4 benchmarks.
Hybrid won on all 4. The reranker won on 3. BM25 tuning gained under 0.01 everywhere.
No single config wins. Every index needs its own decisions.
https://t.co/oCAE0LOnqE
@aitization Maybe. I think the thing is we know there's a window where we can win the first to the market and we're pursing it full steam.
Consuming the latest research, shipping prototypes, and being hyper customer obsessed.
We're working super hard to make something great.