[{"data":1,"prerenderedAt":391},["ShallowReactive",2],{"project-36391655":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":43,"post_text":45,"what":46,"said":76,"maker_in_thread":202,"shape":214,"grid":219,"attempts":353,"same_week":354,"maker_shows":34,"maker_hidden":383,"reader_version":384,"card":385},"36391655","HN Recommend","My recommendation engine for Hacker News","title","Show HN: My recommendation engine for Hacker News","https:\u002F\u002Fhn-recommend.julienc.me","hn-recommend.julienc.me","julien040",1687185957,2023,314,58,135,3109,5,"legend","learning_science","search and recommendation for hacker news","web_app",[24,25,26,27,28],"hacker news","recommendation engine","embeddings","semantic search","faiss","debated",0.013,{"positive":32,"neutral":33,"mixed":34,"critical":35},8,12,3,9,32,{"idea":38,"execution":40,"design":41},{"positive":34,"critical":39},2,{"positive":18,"critical":18},{"positive":42,"critical":39},0,[42,18,18,42,42,18,34,44,44,34,42,34,34,44,42,34,42,18,34,42,42,34,44,42,18,34,34,42,34,18,44,18,34,44,42,44,34,44,44,34],1,"Hi!\nI’m Julien and I built a recommendation engine for Hacker News.\n\nI feel like this website is a gold mine. Every day, I find some very interesting stories about a topic. And sometimes, I want to find other stories covering that same topic but I can’t.\n\nHacker News has years of history of awesome discussion and ressources. Unfortunately, I think HN Algolia isn’t helpful in searching these old threads. As a student, I want to learn a lot from this website.\n\nThis is why I created HN Recommend. Input a sentence or the URL of an article, and get the most popular and similar posts from Hacker New",{"problem":47,"audience":53,"pricing":57,"motivation":58,"time_to_build":57,"stack":62,"openness":73,"stage":74,"team_size":75},{"value":48,"source":49,"evidence":50},"HN Algolia struggles to find older related discussions that don't share the same words.","maker",[51],{"id":52},"post",{"value":54,"source":49,"evidence":55},"Hacker News readers, including students learning from past threads.",[56],{"id":52},null,{"value":59,"source":49,"evidence":60},"Student wanting to learn from years of HN discussions and find related stories.",[61],{"id":52},[63,67],{"value":64,"source":49,"evidence":65},"Faiss",[66],{"id":52},{"value":68,"source":49,"evidence":69},"OpenAI text-embedding-ada-002",[70],{"id":71,"by":11,"text":72},"36397466","Yes it's an issue. Sadly, I can't fix it. I'm using the closed source \"text-embedding-ada-002\" model from OpenAI. As I can see, the longer the input, the more accurate the…","open_source","launched","solo",{"praise":77,"objections":105,"requests":141,"questions":168,"alternatives":183},[78,94],{"value":79,"source":80,"evidence":81},"Useful for finding related posts and expanding topic exposure.","commenter",[82,86,90],{"id":83,"by":84,"text":85},"36396165","dpe82","This is great. I often come across some HN post on a topic I am interested in and then want to go look at other posts in the same topic cluster to expand my exposure. This looks…",{"id":87,"by":88,"text":89},"36423920","RileyJames","Love it. This response is very reactive heavy, where as it’s elixir I’m more interested in. But well done on the execution. It does exactly what it states. I’ve bookmarked. I…",{"id":91,"by":92,"text":93},"36400220","sukki07","This is amazing, thank you for this. Makes finding stuff a lot easier",{"value":95,"source":80,"evidence":96},"Fast turnaround on requested features.",[97,101],{"id":98,"by":99,"text":100},"36394556","dxbydt","Very fast turnaround. Kudos! Works very nicely now. Try this: https:\u002F\u002Fhn-recommend.julienc.me\u002F?q=sf%20crime With the Newest filter vs the Oldest filter. ( btw the default…",{"id":102,"by":103,"text":104},"36397623","TechBro8615","Wow, very fast turnaround time!! I just added it and it works :) Nice job!",[106,125,135],{"value":107,"source":80,"evidence":108},"Search returns unrelated results, especially for single words or short queries.",[109,113,117,121],{"id":110,"by":111,"text":112},"36393449","benzible","Hmm I tried searching \"elixir\" and found nothing related to the language. HN Algolia gives me exactly what I want. On what basis do you say it's \"not helpful\"?",{"id":114,"by":115,"text":116},"36393452","danvayn","hey Julien. I love the product but the search doesn't seem to be doing the best for me. For example, I looked up Tailwind and got plenty of results but none of them actually…",{"id":118,"by":119,"text":120},"36397381","sogen","A comment about search results: \"design system\" is related to design, \"system design\" relates to computing It seems search takes the two inputs as the same. Also, search doesn't…",{"id":122,"by":123,"text":124},"36399097","2h","This URL fails https:\u002F\u002Fhn-recommend.julienc.me\u002F?q=Go",{"value":126,"source":80,"evidence":127},"Does not match exact words, so known posts can be missed.",[128,132],{"id":129,"by":130,"text":131},"36394132","4hEn","I search for a url I know was posted and it doesn't show it. It shows unrelated articles.",{"id":133,"by":130,"text":134},"36395092","It's 10 years old. This search query https:\u002F\u002Fhn-recommend.julienc.me\u002F?q=paul%20graham returns articles that are missing both words of the query",{"value":136,"source":80,"evidence":137},"Results include old posts unless sorted by recency.",[138],{"id":139,"by":99,"text":140},"36393517","Pls sort by recency. Otherwise you see 13 year old articles most of them obsolete\u002Firrelevant to the current situation.",[142,150,154,164],{"value":143,"source":80,"evidence":144},"Add recency or time filter.",[145,146],{"id":139,"by":99,"text":140},{"id":147,"by":148,"text":149},"36393754","lfkdev","A time filter is needed",{"value":151,"source":80,"evidence":152},"Add tagging or keyword filtering.",[153],{"id":114,"by":115,"text":116},{"value":155,"source":80,"evidence":156},"Build a browser extension or mobile shortcut.",[157,161],{"id":158,"by":159,"text":160},"36396428","swyx","i like the idea of this but wont remember it because my muscle memory is tuned to news.ycombinator.com. perhaps i can recommend a chrome extension instead of a website?",{"id":162,"by":103,"text":163},"36396636","An iOS share widget would be cool too. Since you support putting the input text in the URL, then maybe someone can make a Workflow for it and share it here.",{"value":165,"source":80,"evidence":166},"Let users adjust weighting between relevance, score and recency.",[167],{"id":83,"by":84,"text":85},[169,176],{"value":170,"source":80,"evidence":171},"Why single words perform poorly.",[172],{"id":173,"by":174,"text":175},"36408387","Solvency","Out of curiosity related to the word vectorization algorithm...why does one word not perform as well? Whats the cause\u002Frationale?",{"value":177,"source":80,"evidence":178},"How many articles were skipped during embedding.",[179],{"id":180,"by":181,"text":182},"36399697","jimmySixDOF","The blog post link on GitHub was a nice walk through of your method and I was interested in what you think the hit rate was for getting successful text for embeddings from TFA…",[184,191,195],{"value":185,"source":80,"evidence":186},"askhn.ai",[187],{"id":188,"by":189,"text":190},"36396785","lettergram","Checkout https:\u002F\u002Faskhn.ai The content is ranked by how people discuss the topics and who discusses them If you just do embeddings on posts you might miss relevant content. When…",{"value":192,"source":80,"evidence":193},"HN Algolia",[194],{"id":110,"by":111,"text":112},{"value":196,"source":80,"evidence":197},"hackernewsletter.com",[198],{"id":199,"by":200,"text":201},"36397456","ColinWright","You mean like the one that's emailed to me every week? https:\u002F\u002Fhackernewsletter.com",{"comments":203,"answered":35,"key":204},20,[205,208,211],{"id":206,"text":207},"36394261","I just released a new update, thanks to everyone's feedback. Now, you can sort results by relevancy, age, or score using the select.",{"id":209,"text":210},"36396571","I thought about an algorithm with weight adjustable by the user. Now, the API returns a field with the distance between the post and the query (the square of the Euclidean…",{"id":212,"text":213},"36393533","Yes the search doesn't work very well for one word. Try to input an url about elixir like this:…",{"commenters":215,"max_depth":216,"maker_replies":203,"first_comment_mins":217,"lifespan_hours":218},30,7,133,205,[220,224,228,230,234,238,241,243,245,247,250,252,254,256,257,259,262,264,267,270,271,274,275,277,279,280,283,287,289,291,293,294,296,297,298,300,302,304,306,307,309,312,314,315,317,319,323,327,330,332,334,335,337,341,344,346,348,351],[221,42,42,222,223,42],"36398055","other\u002Fneutral\u002Fnone","samwho",[225,44,18,226,227,42],"36398977","project\u002Fpositive\u002Fidea","FieryTransition",[206,39,42,229,11,44],"project\u002Fneutral\u002Fdesign",[231,34,18,232,233,42],"36396477","project\u002Fpositive\u002Fexecution","moritzwarhier",[235,34,42,236,237,42],"36398484","commenter\u002Fneutral\u002Fnone","febed",[239,34,42,236,240,42],"36398632","SkyArrow",[242,34,42,236,233,42],"36408651",[83,244,18,226,84,42],4,[209,244,42,246,11,44],"project\u002Fneutral\u002Fidea",[248,244,34,246,249,42],"36399654","juliusgeo",[188,244,44,251,189,42],"project\u002Fcritical\u002Fidea",[110,18,44,253,111,42],"project\u002Fcritical\u002Fexecution",[212,18,42,255,11,44],"project\u002Fneutral\u002Fexecution",[173,18,34,255,174,42],[258,18,42,255,11,44],"36430855",[114,260,34,261,115,42],6,"project\u002Fmixed\u002Fidea",[263,260,42,246,11,44],"36393901",[265,216,34,246,266,42],"36403583","wseqyrku",[268,32,34,246,269,42],"36396575","akomtu",[199,32,44,251,200,42],[272,32,42,222,273,42],"36398505","balder1991",[139,35,34,246,99,42],[276,35,42,246,11,44],"36393932",[278,35,42,229,11,44],"36394275",[98,35,18,232,99,42],[87,281,34,282,88,42],10,"project\u002Fmixed\u002Fexecution",[284,285,34,246,286,42],"36394441",11,"fewald_net",[288,285,42,246,11,44],"36394811",[290,285,42,222,286,42],"36398288",[292,285,42,236,11,44],"36401274",[180,285,34,255,181,42],[295,285,42,255,11,44],"36401206",[118,33,44,253,119,42],[71,33,42,255,11,44],[299,33,18,232,119,42],"36501878",[91,301,18,232,92,42],13,[158,303,34,246,159,42],14,[305,303,42,246,11,44],"36396600",[162,303,34,246,103,42],[308,303,42,236,11,44],"36396743",[310,303,34,311,103,42],"36396864","maker\u002Fneutral\u002Fnone",[313,303,42,255,11,44],"36397060",[102,303,18,232,103,42],[122,316,44,253,123,42],15,[318,316,42,255,11,44],"36401228",[320,321,18,226,322,42],"36418685",16,"rjrobben",[324,325,34,246,326,42],"36394442",17,"passion__desire",[328,325,42,329,11,44],"36394715","project\u002Fneutral\u002Fprivacy",[129,331,44,253,130,42],18,[333,331,42,255,11,44],"36394200",[133,331,44,253,130,42],[336,331,42,255,11,44],"36395226",[338,339,34,282,340,42],"36395329",19,"rounakdatta",[342,203,44,343,273,42],"36398433","project\u002Fcritical\u002Fdesign",[345,203,42,229,11,44],"36401107",[347,203,44,343,273,42],"36403975",[349,203,42,350,11,44],"36448981","project\u002Fpositive\u002Fdesign",[147,352,34,246,148,42],21,[],[355,360,365,370,375,379],{"id":356,"name":357,"points":358,"category":359},"36365511","PropBox",520,"web_saas",{"id":361,"name":362,"points":363,"category":364},"36360789","Novel",370,"productivity",{"id":366,"name":367,"points":368,"category":369},"36383773","Answer Overflow",333,"data",{"id":371,"name":372,"points":373,"category":374},"36388821","Slint",291,"dev_tools",{"id":376,"name":377,"points":378,"category":374},"36422730","gpt-engineer",178,{"id":380,"name":381,"points":382,"category":369},"36409201","Autolabel",153,false,"v2",{"id":386,"name":5,"line":6,"ai":383,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":387,"top":18,"kw":388,"maker":11,"domain":10,"tier":19,"attempts":42,"strip":389,"url":9,"mix":390,"praise":79,"objection":107,"sub":21},36391655,0.01,[24,25,26,27,28],"0550053113033103053003105330351531013113",[32,33,34,35],1794036637126]