Airweave is an open-source dev tool that lets agents search any app. It connects to apps, databases, or document stores and turns their contents into searchable knowledge bases for agents.
A few months ago we were building agents that interacted with various apps. We were frustrated when they struggled with vague natural language requests like "resolve that Jira ticket about missing auth configs", "refund payments in Stripe for unsatisfied customers", or "what were Q1 returns from the financial sheet in gdrive?" The agents would then inefficiently chain multiple function calls, fail to retrieve data, or hallucinate answers.
We also noticed that despite the rise of MCP creating more desire for agents to interact with external resources, the majority of agent dev tooling focused on function calling and actions instead of search. We were annoyed by the lack of tooling that enabled agents to semantically search workspace or database contents, so we started building Airweave first as an internal solution. Then we decided to open-source it and pursue it full time after we saw the positive reactions from agent builders.
The best part for us about building Airweave is continually seeing the amazing things our users create with it.
Congrats on the launch. Can you tell me more about how the underlying data sync works? Does it run on a schedule? What if there are conflicting data? How “real-time” is the data getting indexed?
thanks @renebrandel! ofc, data is synced either manually or scheduled using cron jobs. event-based updates are a pro-feature that also depend on whether the source's API support webhooks. And we use content hashing for version tracking of inserts, updates, and deletions. Lmk if you have any other questions!
@renebrandel Maturity of the API is really important here. If the API allows easy changelogs/webhooks, it can be near real-time. Some more legacy APIs we can tune to 5 min.
I really appreciate the ability to connect to multiple apps and databases all in one place. It seems like a great way to simplify how developers search through their data and keep everything organized.
thanks @emirkarabeg ! Airweave returns a scored list of result that can be passed to rerankers, so teams can tune how results are ordered before they’re fed to LLMs down the line. or airweave can also apply reranking and recency bias itself. If you’re thinking of another ranking use case, let me know! curious to hear
Airweave is a brilliant open-source solution that empowers developers to transform any app's data into a searchable knowledge base. The ability to connect seamlessly to diverse sources like databases and documents makes building intelligent agents so much more efficient. Excited to see how it evolves!
Airweave sounds like a huge step forward in enabling AI agents to truly understand and interact with data across apps and databases. Can Airweave handle complex queries across multiple apps at once, or is it designed to focus on a single app/database at a time?
@evgenii_zaitsev1 it's specifically designed to be able to manage multiple sources at once! You can define a collection which contains multiple source_connections. That collection you can search :)
Love the concept. Organizing thoughts and research visually makes such a difference, especially when things start to get messy. Definitely something I can see myself using. Big congrats to the team! 🎉
This looks awesome! On the demo it looks like it is faster vs Claude - do you have an estimate on that? Also, what when there are multiple endpoints to look through?
@antoine_bouchardy endlessly chained MCP often end up in hallucination, which gives you nothing, so I guess infinitely faster (lol). Airweave aggregates all the knowledge into one endpoint that the search logic behind then manages
Really cool approach. For teams with a lot of sensitive or regulated data, what’s your advice for balancing searchability with security and access controls? Any best practices you’ve seen work well?
@paul_sangle_ferriere1 in general our users just specify the parts that are not security-constrained. This mostly works out just fine because they don't want agents working on tricky parts anyway :)
The struggle to get agents to accurately retrieve information is real. The frustration of agents struggling with vague natural language requests and inefficiently chaining function calls is such a real pain point.
Congrats on the launch! Could also be interesting to explore integrations with domain-specific platforms (or some way for users to easily build connectors to them) like healthcare EHRs.
Replies
Airweave
A few months ago we were building agents that interacted with various apps. We were frustrated when they struggled with vague natural language requests like "resolve that Jira ticket about missing auth configs", "refund payments in Stripe for unsatisfied customers", or "what were Q1 returns from the financial sheet in gdrive?" The agents would then inefficiently chain multiple function calls, fail to retrieve data, or hallucinate answers.
We also noticed that despite the rise of MCP creating more desire for agents to interact with external resources, the majority of agent dev tooling focused on function calling and actions instead of search. We were annoyed by the lack of tooling that enabled agents to semantically search workspace or database contents, so we started building Airweave first as an internal solution. Then we decided to open-source it and pursue it full time after we saw the positive reactions from agent builders.
The best part for us about building Airweave is continually seeing the amazing things our users create with it.
Looking forward to the comments 👋
- Lennert & Rauf
@lennertjansen Love the focus on search.
Airweave
@lennertjansen @masump thank you!
Airweave
@masump appreciate it!!
Den
You guys are building the most important layer for the next generation of mcp-native applications. 🦾🫡
Airweave
appreciate that @justin_lee27!! 🙏
Airweave
@justin_lee27 Love working with you team!!
AWS Amplify
Congrats on the launch. Can you tell me more about how the underlying data sync works? Does it run on a schedule? What if there are conflicting data? How “real-time” is the data getting indexed?
Airweave
thanks @renebrandel! ofc, data is synced either manually or scheduled using cron jobs. event-based updates are a pro-feature that also depend on whether the source's API support webhooks. And we use content hashing for version tracking of inserts, updates, and deletions. Lmk if you have any other questions!
Airweave
@renebrandel Maturity of the API is really important here. If the API allows easy changelogs/webhooks, it can be near real-time. Some more legacy APIs we can tune to 5 min.
I really appreciate the ability to connect to multiple apps and databases all in one place. It seems like a great way to simplify how developers search through their data and keep everything organized.
Airweave
thanks @dick_carville, great to hear that 🙂
Airweave
@dick_carville Thanks a lot!
Sim Studio
Congrats on the launch! Your product looks really cool. Do you see this as a ranking opportunity for startups as well with LLM responses?
Airweave
thanks @emirkarabeg ! Airweave returns a scored list of result that can be passed to rerankers, so teams can tune how results are ordered before they’re fed to LLMs down the line. or airweave can also apply reranking and recency bias itself. If you’re thinking of another ranking use case, let me know! curious to hear
Airweave is a brilliant open-source solution that empowers developers to transform any app's data into a searchable knowledge base. The ability to connect seamlessly to diverse sources like databases and documents makes building intelligent agents so much more efficient. Excited to see how it evolves!
Airweave
thank you@supa_l, really appreciate you saying that!
Airweave
@supa_l Thanks a lot!
Den
airweave has been an amazing product, thanks both!!
Airweave
thanks @linus_talacko1 🫡🫡🫡🫡
Congrats! The looks awesome. Curious do you see this as a chance for startups to improve their rankings through LLM-generated responses too?
Airweave
@bereket_engida we keep getting surprised by how the community uses Airweave. That definitely sounds plausible as well
Airweave sounds like a huge step forward in enabling AI agents to truly understand and interact with data across apps and databases. Can Airweave handle complex queries across multiple apps at once, or is it designed to focus on a single app/database at a time?
Airweave
@evgenii_zaitsev1 it's specifically designed to be able to manage multiple sources at once! You can define a collection which contains multiple source_connections. That collection you can search :)
Love the concept. Organizing thoughts and research visually makes such a difference, especially when things start to get messy. Definitely something I can see myself using. Big congrats to the team! 🎉
Airweave
@giga_chkhikvadze1 hey, happy to help when you give it a spin!
Dash
Great team, and great product!
Airweave
@harsha_gaddipati hey, right back at you!
Probo
This looks awesome! On the demo it looks like it is faster vs Claude - do you have an estimate on that? Also, what when there are multiple endpoints to look through?
Airweave
@antoine_bouchardy endlessly chained MCP often end up in hallucination, which gives you nothing, so I guess infinitely faster (lol). Airweave aggregates all the knowledge into one endpoint that the search logic behind then manages
cubic
Really cool approach. For teams with a lot of sensitive or regulated data, what’s your advice for balancing searchability with security and access controls? Any best practices you’ve seen work well?
Airweave
@paul_sangle_ferriere1 in general our users just specify the parts that are not security-constrained. This mostly works out just fine because they don't want agents working on tricky parts anyway :)
Dash
We use airweave, and its great!
Airweave
great to be working together man @dhruv_roongta !
I love the fact that you built this out of your own needs.
that's super useful
Morphik
It's so good, even works better than the SOTA providers' direct integrations.
Yes that will definitely be needed. Congrats on the launch and will be tracking progress, all the best!
Jazzberry
The struggle to get agents to accurately retrieve information is real. The frustration of agents struggling with vague natural language requests and inefficiently chaining function calls is such a real pain point.
Congrats on the launch! Could also be interesting to explore integrations with domain-specific platforms (or some way for users to easily build connectors to them) like healthcare EHRs.