[{"data":1,"prerenderedAt":245},["ShallowReactive",2],{"project-41236439":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":82,"maker_in_thread":142,"shape":155,"grid":159,"attempts":205,"same_week":206,"maker_shows":32,"maker_hidden":237,"reader_version":238,"card":239},"41236439","H2LooP Studio","AI co-worker for system software development (Rust,C,C++,pdf)","title","Show HN: AI co-worker for system software development (Rust,C,C++,pdf)","https:\u002F\u002Fh2loop.ai\u002F","h2loop.ai","yosai",1723563147,2024,22,19,1827,3515,52,"shown","dev_tools","ai coding assistant","web_app",[24,25,26,27,28],"system software","embedded code generation","technical specifications","datasheet parsing","fine-tuned coding model","curious",0.167,{"positive":32,"neutral":33,"mixed":34,"critical":35},2,5,1,0,8,{"idea":38,"performance":39,"execution":40,"docs":41,"privacy":42},{"positive":34,"critical":35},{"positive":34,"critical":35},{"positive":35,"critical":35},{"positive":35,"critical":35},{"positive":35,"critical":35},[44,35,35,35,33,35,35,35,33,35,44,35,35,44,35,44,44,44,35],3,"Hey Everybody,\n\nWe are really excited to release the 1st version of H2LooP studio today.https:\u002F\u002Fh2loop.ai\u002F H2LooP Studio helps system software engineers generate code from technical specs, debug issues, and understand complex code in C, C++, Go, and Rust. Under the hood, it uses the H2LooP Data Engine to create instruction-tuned datasets from data sheets and source code.\n\nModels are what they eat. We create high-quality, pre-vetted domain-specific training data (telecom, IoT, automotive, consumer electronics) at scale for fine-tuning small language models. We leverage both LLMs and human exper",{"problem":47,"audience":53,"pricing":57,"motivation":58,"time_to_build":57,"stack":62,"openness":79,"stage":80,"team_size":81},{"value":48,"source":49,"evidence":50},"Generic coding assistants fail at system software tasks like driver code and reading PDF technical specs","maker",[51],{"id":52},"post",{"value":54,"source":49,"evidence":55},"System software engineers, developers, product managers and CTOs working on proprietary software",[56],{"id":52},null,{"value":59,"source":49,"evidence":60},"Generic coding assistants cannot handle technical specifications or crash logs needed for system software",[61],{"id":52},[63,67,71,75],{"value":64,"source":49,"evidence":65},"C",[66],{"id":52},{"value":68,"source":49,"evidence":69},"C++",[70],{"id":52},{"value":72,"source":49,"evidence":73},"Go",[74],{"id":52},{"value":76,"source":49,"evidence":77},"Rust",[78],{"id":52},"unknown","beta","team",{"praise":83,"objections":99,"requests":107,"questions":114,"alternatives":129},[84,92],{"value":85,"source":86,"evidence":87},"Expert review and validation adds reliability missing in generic assistants","commenter",[88],{"id":89,"by":90,"text":91},"41236729","shrianshag","This is an exciting development for system software engineers working with complex code in languages like C, C++, Go, and Rust. The integration of expert review and validation…",{"value":93,"source":86,"evidence":94},"Useful for OS dev and summarising hardware specs",[95],{"id":96,"by":97,"text":98},"41250040","sim7c00","very cool! ive been using chatgpt for OS dev, to summarize specs and try to make bulletpoints for stuff todo etc. but its a nightmare :'). 100k documentation pages is a pain to…",[100],{"value":101,"source":86,"evidence":102},"Post text appears AI generated",[103],{"id":104,"by":105,"text":106},"41268658","ramon156","I don't want to be that dude, but the tect seems AI generated. Cool product though! I'll give it a go",[108],{"value":109,"source":86,"evidence":110},"Self-hosted version",[111],{"id":112,"by":90,"text":113},"41272265","Will there be a self-hosted version of H2LooP Studio in the future, or will it remain cloud-only?",[115,122],{"value":116,"source":86,"evidence":117},"How expert review and dataset quality are ensured",[118],{"id":119,"by":120,"text":121},"41236582","novacode007","Looks interesting to me, but I had some questions. Could you elaborate on the process of expert review and validation? How do you ensure the quality and accuracy of the datasets…",{"value":123,"source":86,"evidence":124},"How it fares against Devin and Cognition",[125],{"id":126,"by":127,"text":128},"41237435","rnavi",".. and how does it fare with devin\u002Fcognition etc - have not tried them just youtube videos.",[130,134,138],{"value":131,"source":86,"evidence":132},"GitHub Copilot",[133],{"id":52},{"value":135,"source":86,"evidence":136},"ChatGPT",[137],{"id":96,"by":97,"text":98},{"value":139,"source":86,"evidence":140},"Devin and Cognition",[141],{"id":126,"by":127,"text":128},{"comments":143,"answered":144,"key":145},10,4,[146,149,152],{"id":147,"text":148},"41237091","Yes rnavi,I have the same experience as well.The complete system software development can be expedited using AI.Yes, we are doing private beta with a Japanese semiconductor…",{"id":150,"text":151},"41237614","Devin and Cognition are great as co-workers for application software. They're proficient in JavaScript and Python and excel at solving logical problems in software. However, they…",{"id":153,"text":154},"41236762","Thanks shrianshag..You can try some sample Datasheets(pdf) and ask code generation questions.",{"commenters":156,"max_depth":144,"maker_replies":143,"first_comment_mins":157,"lifespan_hours":158},7,16,86,[160,163,165,167,169,171,173,175,177,179,182,184,187,189,192,195,197,200,203],[161,35,44,162,127,35],"41237033","maker\u002Fneutral\u002Fnone",[147,35,35,164,11,34],"project\u002Fpositive\u002Fprivacy",[126,35,35,166,127,35],"commenter\u002Fneutral\u002Fnone",[150,35,35,168,11,34],"project\u002Fneutral\u002Fidea",[89,34,33,170,90,35],"project\u002Fpositive\u002Fidea",[153,34,35,172,11,34],"maker\u002Fneutral\u002Fdocs",[174,34,35,172,11,34],"41236869",[176,32,35,172,11,34],"41236728",[96,44,33,178,97,35],"project\u002Fpositive\u002Fperformance",[180,44,35,181,11,34],"41256535","maker\u002Fneutral\u002Fidea",[119,144,44,183,120,35],"maker\u002Fneutral\u002Fexecution",[185,144,35,186,11,34],"41236682","project\u002Fneutral\u002Fexecution",[188,144,35,172,11,34],"41236734",[190,144,44,191,120,35],"41236754","project\u002Fneutral\u002Fperformance",[193,144,35,194,11,34],"41236782","project\u002Fmixed\u002Fexecution",[104,33,44,196,105,35],"project\u002Fmixed\u002Fdocs",[112,198,44,199,90,35],6,"project\u002Fneutral\u002Fprivacy",[201,156,44,191,202,35],"41237171","cjtechie",[204,156,35,164,11,34],"41237250",[],[207,212,217,222,227,232],{"id":208,"name":209,"points":210,"category":211},"41247023","ytch",2741,"consumer_social",{"id":213,"name":214,"points":215,"category":216},"41224689","PGlite",509,"data",{"id":218,"name":219,"points":220,"category":221},"41203306","LLM-aided OCR",479,"productivity",{"id":223,"name":224,"points":225,"category":226},"41217109","Grandma's Word",454,"games",{"id":228,"name":229,"points":230,"category":231},"41203475","Thunder Compute",344,"infrastructure",{"id":233,"name":234,"points":235,"category":236},"41213711","flutter_rust_bridge",301,"languages",false,"v2",{"id":240,"name":5,"line":6,"ai":237,"cat":20,"year":13,"points":14,"comments":15,"rec":29,"s":241,"top":18,"kw":242,"maker":11,"domain":10,"tier":19,"attempts":35,"strip":243,"url":9,"mix":244,"praise":85,"objection":101,"sub":21},41236439,0.17,[24,25,26,27,28],"3000500050300303330",[32,33,34,35],1793991097864]