[{"data":1,"prerenderedAt":360},["ShallowReactive",2],{"project-43964201":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":10,"maker":11,"posted_at":12,"year":13,"points":14,"comments":15,"rank_in_year":16,"shown_in_year":17,"top":18,"tier":19,"category":20,"subcategory":21,"format":22,"keywords":23,"reception":29,"sentiment":30,"stance_mix":31,"scored_comments":36,"aspects":37,"strip":47,"post_text":48,"what":49,"said":79,"maker_in_thread":199,"shape":202,"grid":207,"attempts":319,"same_week":324,"maker_shows":34,"maker_hidden":352,"reader_version":353,"card":354},"43964201","Airweave","Let agents search any app","title","Show HN: Airweave – Let agents search any app","https:\u002F\u002Fgithub.com\u002Fairweave-ai\u002Fairweave","github.com","lennertjansen",1747064061,2025,176,40,386,3837,11,"notable","ai","agent search infrastructure","library",[24,25,26,27,28],"agent search","mcp server","rag","semantic search","data connectors","curious",0.115,{"positive":32,"neutral":33,"mixed":34,"critical":35},5,27,1,2,35,{"idea":38,"design":40,"performance":42,"execution":43,"privacy":44,"pricing":45,"docs":46},{"positive":39,"critical":35},3,{"positive":41,"critical":41},0,{"positive":34,"critical":41},{"positive":41,"critical":41},{"positive":41,"critical":41},{"positive":41,"critical":41},{"positive":41,"critical":41},[34,41,39,39,39,39,39,39,41,32,39,39,39,32,39,39,39,39,39,39,32,39,39,39,39,39,39,39,39,39,39,39,41,39,32,39,34,41,41,32],"Hey HN, we're Lennert and Rauf. We’re building Airweave (https:\u002F\u002Fgithub.com\u002Fairweave-ai\u002Fairweave), an open-source tool that lets agents search and retrieve data from any app or database. Here’s a general intro: https:\u002F\u002Fwww.youtube.com\u002Fwatch?v=EFI-7SYGQ48, and here’s a longer one that shows more real-world use cases, examples of how Airweave is used by Cursor (0:33) and Claude desktop (2:04), etc.: https:\u002F\u002Fyoutu.be\u002Fp2dl-39HwQo\n\nA couple of months ago we were building agents that interacted with different apps and were frustrated when they struggled to handle vague natural language requests like",{"problem":50,"audience":56,"pricing":64,"motivation":70,"time_to_build":74,"stack":75,"openness":76,"stage":77,"team_size":78},{"value":51,"source":52,"evidence":53},"Agents struggle to find the right data in apps and databases from vague natural language requests","maker",[54],{"id":55},"post",{"value":57,"source":52,"evidence":58},"Developers building AI agents that need access to user or company data",[59,60],{"id":55},{"id":61,"by":62,"text":63},"43965350","raufakdemir","Co-founder here. The Airweave interface doesn't discriminate which downstream use case it's applied in. Most current developers don't build it for a chat interface at all…",{"value":65,"source":52,"evidence":66},"Custom deployments for enterprise customers; a managed service on a flat fee subscription is being released",[67],{"id":68,"by":11,"text":69},"43966274","hi cofounder here. until now it's been custom deployments for customers with additional b2b\u002Fenterprise features. we're also releasing a managed service for a flat fee subscription",{"value":71,"source":52,"evidence":72},"Agents hallucinated or chained many function calls to find data; existing tooling focused on actions rather than search",[73],{"id":55},null,[],"open_source","launched","team",{"praise":80,"objections":93,"requests":118,"questions":137,"alternatives":173},[81,89],{"value":82,"source":83,"evidence":84},"Retrieval accuracy better than other MCP server providers","commenter",[85],{"id":86,"by":87,"text":88},"43964832","renesultan","Had meetings with a ton of MCP-server providers, no one came close to Airweave’s retrieval accuracy. I even tried Zapier and similar large companies, didn’t come near airweave.…",{"value":90,"source":83,"evidence":91},"Team is responsive and supportive",[92],{"id":86,"by":87,"text":88},[94,101,112],{"value":95,"source":83,"evidence":96},"Connector business is a slog with many failures",[97],{"id":98,"by":99,"text":100},"43970052","swyx","i think before you guys spend the next 2 years building this startup, you should carefully study the connector business and the many carcasses along the way. YC itself has a few.…",{"value":102,"source":83,"evidence":103},"Copies customer data and runs RAG over it, raising privacy concerns",[104,108],{"id":105,"by":106,"text":107},"43968838","thawab","Yes, a lot MCP servers are just api wrappers. Airweave looks like it copies the data and has a RAG that is processing your queries.",{"id":109,"by":110,"text":111},"43967366","mike_d","How do you handle data retention? For example say that you suck in the information of a California resident and the company is obligated by law to delete it on request. How do…",{"value":113,"source":83,"evidence":114},"Role-based access control is not yet built",[115],{"id":116,"by":62,"text":117},"43965804","I assume you're talking about the data layer (not the control plane)? We are currently in PoC phase for mapping the role graphs from source systems (Asana, Google Drive) to our…",[119,126,133],{"value":120,"source":83,"evidence":121},"Support for Snowflake and Fabric data lakes",[122],{"id":123,"by":124,"text":125},"43964877","om0agarwal","Looks like a great product! Do you integrate with datalakes (Snowflake\u002FFabric?)",{"value":127,"source":83,"evidence":128},"Ability to take actions such as creating tickets or sending messages",[129],{"id":130,"by":131,"text":132},"43969681","andes314","How about acting? Can your platform have a chat window create a ticket or send a message? I feel there is so much search already.",{"value":134,"source":83,"evidence":135},"Deletion of data by user identifier for compliance",[136],{"id":109,"by":110,"text":111},[138,145,152,159,166],{"value":139,"source":83,"evidence":140},"How is it different from regular MCP servers",[141],{"id":142,"by":143,"text":144},"43965128","bartjanjorna","How is this different from regular MCP servers?",{"value":146,"source":83,"evidence":147},"How the LLM is made aware of the data store beyond a generic search function",[148],{"id":149,"by":150,"text":151},"43969084","smattiso","This is a great idea. I have a question: Typically speaking an LLM is the code driving the control flow and the MCP servers are kind of dumb API endpoints (find_flights,…",{"value":153,"source":83,"evidence":154},"Whether it can run inside a VPC and connect to private databases",[155],{"id":156,"by":157,"text":158},"43965355","brene","Pretty cool stuff. How does it deal with self-hosted data sources? can it run inside a VPC and talk to my RDS instances directly?",{"value":160,"source":83,"evidence":161},"How the product makes money and where data is stored",[162],{"id":163,"by":164,"text":165},"43974248","beecasthurlbow","Looks really cool! I had slight trouble understanding whether the repo is the complete codebase or if it connects to a separate backend. Does one host the server or is it…",{"value":167,"source":83,"evidence":168},"How incremental sync and diffing work across integrations",[169],{"id":170,"by":171,"text":172},"43968059","throwaway314155","Are integrations hooked into via their MCP implementation? Or are you hooking in more traditionally and then exposing MCP on top of that? Also, are these one-time\u002Fevent-based…",[174,181,185,192],{"value":175,"source":83,"evidence":176},"Glean, Onyx, Moveworks, Dashworks",[177],{"id":178,"by":179,"text":180},"43965140","valianter","Is chat always the best interface for all of these apps? I feel like search is the natural first step, but chat-based search has been around for a while. Feel like an MCP-based…",{"value":182,"source":83,"evidence":183},"Zapier",[184],{"id":86,"by":87,"text":88},{"value":186,"source":83,"evidence":187},"n8n",[188],{"id":189,"by":190,"text":191},"43965533","pomarie","Pretty cool – when does it make sense to use this vs n8n?",{"value":193,"source":83,"evidence":194},"AWS Bedrock knowledge bases",[195],{"id":196,"by":197,"text":198},"43968457","rohanv23","Used to work at AWS bedrock knowledge bases — love the way yall are thinking about this can't wait to try it out.",{"comments":34,"answered":34,"key":200},[201],{"id":68,"text":69},{"commenters":203,"max_depth":204,"maker_replies":34,"first_comment_mins":205,"lifespan_hours":206},29,4,56,141,[208,210,214,216,220,222,224,226,229,233,235,238,240,242,247,249,252,255,257,259,261,262,267,269,272,274,276,278,282,283,284,287,291,293,297,301,303,308,311,314],[98,41,34,209,99,41],"maker\u002Fcritical\u002Fidea",[211,41,41,212,213,41],"44020962","commenter\u002Fneutral\u002Fidea","topaztee",[149,34,39,215,150,41],"project\u002Fneutral\u002Fdesign",[217,34,39,218,219,41],"43969337","project\u002Fmixed\u002Fperformance","behindsight",[178,35,39,221,179,41],"project\u002Fneutral\u002Fidea",[61,35,39,223,62,41],"maker\u002Fneutral\u002Fidea",[170,39,39,225,171,41],"project\u002Fneutral\u002Fexecution",[227,39,39,228,62,41],"43968172","maker\u002Fneutral\u002Fexecution",[230,204,41,231,232,41],"43994157","other\u002Fneutral\u002Fnone","alephnan",[86,32,32,234,87,41],"project\u002Fpositive\u002Fperformance",[236,32,39,221,237,41],"43966941","risyachka",[239,32,39,221,62,41],"43968182",[105,32,39,241,106,41],"project\u002Fneutral\u002Fprivacy",[243,244,32,245,246,41],"43965020",6,"project\u002Fpositive\u002Fidea","ayxliu",[156,248,39,241,157,41],7,[250,248,39,251,62,41],"43965425","maker\u002Fneutral\u002Fnone",[163,253,39,254,164,41],8,"project\u002Fneutral\u002Fpricing",[130,256,39,221,131,41],9,[142,258,39,221,143,41],10,[260,258,39,223,62,41],"43965281",[196,18,32,245,197,41],[263,264,39,265,266,41],"43966883",12,"project\u002Fneutral\u002Fdocs","howmayiannoyyou",[268,264,39,251,62,41],"43968204",[123,270,39,271,124,41],13,"project\u002Fneutral\u002Fnone",[273,270,39,251,62,41],"43965111",[189,275,39,221,190,41],14,[277,275,39,221,62,41],"43965681",[279,280,39,241,281,41],"43965553",15,"nishanthooda",[116,280,39,241,62,41],[109,280,39,241,110,41],[285,280,39,286,62,41],"43968194","maker\u002Fneutral\u002Fprivacy",[288,289,39,254,290,41],"43966066",16,"modelorona",[68,289,41,292,11,34],"maker\u002Fneutral\u002Fpricing",[294,295,39,271,296,41],"43970219",17,"dr_kiszonka",[298,299,32,245,300,41],"43964909",18,"rajit",[302,299,39,221,62,41],"43965531",[304,305,34,306,307,41],"43965940",19,"project\u002Fcritical\u002Fidea","flockonus",[309,305,41,310,62,41],"43966029","commenter\u002Fneutral\u002Fnone",[312,305,41,231,313,41],"43966671","1317",[315,316,32,317,318,41],"44003027",20,"maker\u002Fpositive\u002Fnone","hommes-r",[320,323],{"id":321,"t":322,"points":35,"comments":35},"42936521",1738693504,{"id":4,"t":12,"points":14,"comments":15},[325,330,335,340,344,349],{"id":326,"name":327,"points":328,"category":329},"43935394","Hyvector",331,"dev_tools",{"id":331,"name":332,"points":333,"category":334},"43928118","qtap",259,"infrastructure",{"id":336,"name":337,"points":338,"category":339},"43975423","HelixDB",237,"data",{"id":341,"name":342,"points":343,"category":20},"43988381","Muscle-Mem",226,{"id":345,"name":346,"points":347,"category":348},"43936097","Aberdeen",224,"languages",{"id":350,"name":351,"points":14,"category":20},"43988533","Semantic Calculator",false,"v2",{"id":355,"name":5,"line":6,"ai":352,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":356,"top":18,"kw":357,"maker":11,"domain":10,"tier":19,"attempts":35,"strip":358,"url":9,"mix":359,"praise":82,"objection":95,"sub":21},43964201,0.12,[24,25,26,27,28],"1033333305333533333353333333333303531005",[32,33,34,35],1794036674675]