Launched this week
Weavable
Give every AI agent persistent work context
375 followers
Give every AI agent persistent work context
375 followers
Weavable gives AI agents persistent, live work context from the tools your business already runs on. Through a single MCP endpoint, it turns scattered updates, relationships, and system changes into a usable context layer so agents can reason more accurately without constantly re-ingesting data. The result is lower token usage, better outputs, and more reliable agent behavior across real business workflows.
Sharing context between different agents is probably a lot more difficult in practice than most people think, especially when synchronization and consistency issues arise in distributed systems. I wonder how Weavable deals with rollback or recovery mechanisms if one integration sends the wrong state updates.
The 90% token reduction vs raw mcp calls is a bold claim but the reasoning makes sense , pre-indexing relationships means the agent isn't burning context figuring out that "Acme Inc" in hubspot is the same entity as "acme.com" in stripe. Curious how stale the context layer gets between syncs, especially for workflows that need near-real-time state like live deal updates or ticket escalations.
Token usage and repeated context loading have honestly been one of the biggest frustrations when working with AI workflows ,so this direction immediately made sense to me. Curious though , how does Weavable handle situations where the same information exists across multiple tools but they contradict each other? Like if HubSpot says a deal is open but a Slack thread suggests it was verbally closed — which source does the agent actually reason from?
okay so I've been trying to build a small agent project for college that pulls from Notion and Gmail and the amount of times it just confidently answered something wrong because it was working off old data was genuinely embarrassing 😭 the RAG vs activity graph distinction makes so much more sense to me now and i think RAG is basically a frozen photo while this is more like a live feed, right ? also curious that when an agent queries Weavable mid-task and the context has literally just updated like seconds ago, does it get the new version or could there still be a small lag?
been tinkering with a lot of mcp lately and love that a product is using a mcp first approach. would definitely love to know more about your approach towards long term context memory over task specific memory. great job guys!
Context loss between sessions is so annoying and nobody really talks about it. Every time you start a new agent run you're basically starting from scratch. If this actually solves persistent context across tools, that's genuinely useful. Skeptical but intrigued. Will definitely try it.
Epsilla (YC S23)
Congratulations. And happy product launch. @abesh_thakur
Weavable
Thanks so much for your support @huisong_li!
Weavable
@huisong_li Thanks for your support - means a lot to us! What kind of AI powered workflows have you been getting the most mileage out of for your day to day?