[{"data":1,"prerenderedAt":295},["ShallowReactive",2],{"project-37433218":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":35,"aspects":36,"strip":41,"post_text":44,"what":45,"said":78,"maker_in_thread":179,"shape":189,"grid":192,"attempts":259,"same_week":260,"maker_shows":40,"maker_hidden":287,"reader_version":288,"card":289},"37433218","Rivet","open-source AI Agent dev env with real-world applications","title","Show HN: Rivet – open-source AI Agent dev env with real-world applications","https:\u002F\u002Frivet.ironcladapp.com\u002F","rivet.ironcladapp.com","gogwilt",1694179745,2023,176,30,335,3109,11,"notable","ai","agent development environment","library",[24,25,26,27,28],"ai agents","visual programming","llm debugging","langchain alternative","typescript llm","loved",0.343,{"positive":32,"neutral":33,"mixed":34,"critical":34},6,9,0,15,{"idea":37,"design":39},{"positive":38,"critical":34},2,{"positive":40,"critical":34},1,[42,34,42,34,42,42,34,34,43,34,42,34,43,34,42,34,42,34,42,34,43,34,43,34,43,34,43,34,42,34],3,5,"We just launched Rivet, the open-source visual AI programming environment! We built Rivet, because we were building complex AI Agent applications at Ironclad. It unlocked our abilities here, and we're excited to make available to the entire community.\n\nBackstory: A few months ago, inspired by things like LangChain and LlamaIndex, we started building an AI agent that could work with legal contracts. Unfortunately, we couldn't just use retrieval augmented generation (RAG), because a lot of contracts are basically identical (many chunks with near-identical embeddings), except for a few key detail",{"problem":46,"audience":52,"pricing":61,"motivation":62,"time_to_build":61,"stack":66,"openness":75,"stage":76,"team_size":77},{"value":47,"source":48,"evidence":49},"Complex AI agent applications are hard to debug and change; visual programming with a remote debugger helps","maker",[50],{"id":51},"post",{"value":53,"source":48,"evidence":54},"Teams building tool-using LLM agent applications, early to growth-stage startups",[55,58],{"id":56,"by":11,"text":57},"37435116","I think it's people building tool-using agent applications. We've been collaborating with several amazing teams over the past few months, who have been pushing Rivet in various…",{"id":59,"by":11,"text":60},"37436977","Likely early to growth-stage start-ups! That's at least what a lot of our early design partners and collaborators have been. Rivet slots in neatly to most modern TypeScript…",null,{"value":63,"source":48,"evidence":64},"Built to solve debugging and iteration pain while building an AI agent for legal contracts at Ironclad",[65],{"id":51},[67,71],{"value":68,"source":48,"evidence":69},"TypeScript",[70],{"id":51},{"value":72,"source":48,"evidence":73},"Python (via a way to run it)",[74],{"id":51},"open_source","launched","company",{"praise":79,"objections":104,"requests":115,"questions":134,"alternatives":159},[80,92,100],{"value":81,"source":82,"evidence":83},"Visual programming is a good fit for LLM work and makes prompting easier","commenter",[84,88],{"id":85,"by":86,"text":87},"37435359","abdellah123","This is a game changer. I have implemented AI features recently, and it was a nightmare using code. Visual programming is a perfect match for AI. And Rivet is very neat and…",{"id":89,"by":90,"text":91},"37434031","bbischof","Super interesting. I haven’t seen much chain support tech for ts folk which is a hard requirement for more production applications.",{"value":93,"source":82,"evidence":94},"Debugger and TypeScript support praised",[95,99],{"id":96,"by":97,"text":98},"37438533","jamesmcintyre","This is awesome! I've been looking for exactly this type of solution that gives a more intuitive (yet robust) ui and has node\u002Ftypescript api (rivets node sdk is even better!).…",{"id":89,"by":90,"text":91},{"value":101,"source":82,"evidence":102},"Polished and neat UI",[103],{"id":85,"by":86,"text":87},[105],{"value":106,"source":48,"evidence":107},"Saving project files is janky, downloads the project file in web mode",[108,112],{"id":109,"by":110,"text":111},"37435947","gsuuon","Looks interesting! Since it's already typescript, are there any plans for a web app in addition to desktop?",{"id":113,"by":11,"text":114},"37436173","Thanks! Actually, when running locally, you can open it in a web app. There's one catch, though: saving project files is pretty janky (it downloads the project file). I think…",[116,123,127],{"value":117,"source":82,"evidence":118},"Support for open-source LLMs such as LocalAI or vLLM",[119],{"id":120,"by":121,"text":122},"37434308","virtuallyvivek","Excited to give this a try! A visual environment seems like a natural fit for chained LLM calls. Have you considered supporting OSS LLMs? Inference servers like LocalAI or vLLM…",{"value":124,"source":82,"evidence":125},"A web app version alongside desktop",[126],{"id":109,"by":110,"text":111},{"value":128,"source":82,"evidence":129},"Real-world open-source example projects",[130],{"id":131,"by":132,"text":133},"37437705","jayfk","Are there any real world projects which are open source using this? Would love to take a peek on how this looks like on something non trivial.",[135,142,149],{"value":136,"source":82,"evidence":137},"How does it choose between plugins, and can an LLM choose them dynamically?",[138,139],{"id":96,"by":97,"text":98},{"id":140,"by":11,"text":141},"37438615","Thanks! Really glad to hear it, and excited to hear what you think! Rivet doesn't have a built-in way of choosing between different ChatGPT plugins, so you have to explicitly…",{"value":143,"source":82,"evidence":144},"How does it compare to langflow and flowise?",[145],{"id":146,"by":147,"text":148},"37438790","denidoman","Nice app, and thank you for OSS! Am I right, that the following 2 services solve the same product in a similar way: https:\u002F\u002Fgithub.com\u002Flogspace-ai\u002Flangflow ,…",{"value":150,"source":82,"evidence":151},"Advice on how many tools to give an agent",[152,156],{"id":153,"by":154,"text":155},"37436243","dsaffy","Awesome work Cai and Andy. Do you have advice on structuring functions for best results? (eg, is it a good idea to give my agent access to things like a calculator \"in case it…",{"id":157,"by":11,"text":158},"37436478","Thanks! I don't think it's a good idea to give an agent tools \"just in case.\" We've opted for giving our agents access to a few distinct, but powerful tools, and then trusting…",[160,164,168,175],{"value":161,"source":82,"evidence":162},"Langflow",[163],{"id":146,"by":147,"text":148},{"value":165,"source":82,"evidence":166},"Flowise",[167],{"id":146,"by":147,"text":148},{"value":169,"source":82,"evidence":170},"LLMStack",[171],{"id":172,"by":173,"text":174},"37434407","ajhai","We recently opensourced a similar platform for building workflows by chaining LLMs visually along with LocalAI support. Check it out at https:\u002F\u002Fgithub.com\u002Ftrypromptly\u002FLLMStack.…",{"value":176,"source":48,"evidence":177},"LangChain and LlamaIndex",[178],{"id":51},{"comments":35,"answered":180,"key":181},8,[182,185,186],{"id":183,"text":184},"37439239","Thanks! Yes, I think Rivet, langflow, and flowise all came to the idea of visual programming for LLMs in parallel. I like to think it's a good sign that visual programming is a…",{"id":140,"text":141},{"id":187,"text":188},"37434463","Nice! Love this!",{"commenters":190,"max_depth":42,"maker_replies":35,"first_comment_mins":191,"lifespan_hours":18},16,66,[193,195,197,198,200,201,202,204,206,208,211,214,215,216,218,221,222,225,228,230,231,234,236,241,243,246,248,252,254,258],[146,34,42,194,147,34],"project\u002Fneutral\u002Fidea",[183,34,34,196,11,40],"project\u002Fpositive\u002Fperformance",[96,40,42,194,97,34],[140,40,34,199,11,40],"project\u002Fneutral\u002Fdocs",[120,38,42,194,121,34],[172,38,42,194,173,34],[187,38,34,203,11,40],"maker\u002Fpositive\u002Fnone",[205,38,34,194,11,40],"37434451",[89,42,43,207,90,34],"project\u002Fpositive\u002Fidea",[209,42,34,210,11,40],"37434205","project\u002Fpositive\u002Fdesign",[153,212,42,213,154,34],4,"maker\u002Fneutral\u002Fidea",[157,212,34,213,11,40],[85,43,43,210,86,34],[217,43,34,203,11,40],"37436195",[219,32,42,194,220,34],"37434952","ankrgyl",[56,32,34,194,11,40],[131,223,42,224,132,34],7,"project\u002Fneutral\u002Fnone",[226,223,34,227,11,40],"37437750","maker\u002Fneutral\u002Fnone",[109,180,42,229,110,34],"project\u002Fneutral\u002Fdesign",[113,180,34,229,11,40],[232,33,43,207,233,34],"37440605","emwang",[235,33,34,203,11,40],"37440951",[237,238,43,239,240,34],"37438141",10,"project\u002Fpositive\u002Fnone","ryscheng",[242,238,34,227,11,40],"37438258",[244,18,43,239,245,34],"37437697","swimburger",[247,18,34,203,11,40],"37437710",[249,250,43,203,251,34],"37435463",12,"mikeknoop",[253,250,34,203,11,40],"37435521",[255,256,42,224,257,34],"37436642",13,"damndan",[59,256,34,224,11,40],[],[261,264,269,274,279,283],{"id":262,"name":263,"points":16,"category":20},"37427127","HackYourNews",{"id":265,"name":266,"points":267,"category":268},"37470611","splat",309,"languages",{"id":270,"name":271,"points":272,"category":273},"37444683","Ghidra Plays Mario",207,"dev_tools",{"id":275,"name":276,"points":277,"category":278},"37499375","Lantern",196,"data",{"id":280,"name":281,"points":282,"category":268},"37438028","cursed-typescript",186,{"id":284,"name":285,"points":286,"category":268},"37448914","view.py",173,false,"v2",{"id":290,"name":5,"line":6,"ai":287,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":291,"top":18,"kw":292,"maker":11,"domain":10,"tier":19,"attempts":34,"strip":293,"url":9,"mix":294,"praise":81,"objection":106,"sub":21},37433218,0.34,[24,25,26,27,28],"303033005030503030305050505030",[32,33,34,34],1794036641495]