{"slug": "micrograd", "name": "micrograd", "category": "dev-libs", "category_name": "Libraries and frameworks", "description": "", "recommenders": 3, "commenters": 5, "warners": 0, "score": 3.159, "rising": 0.35, "first": "2026-01-01", "latest": "2026-08-11", "why": "Take a look at Karpathy's micrograd repo to become more familiar with the process.", "why_is_quote": true, "has_verdict": false, "sentiment": {"recommend": 3, "neutral": 1, "compare": 1, "warn": 0}, "spark": [0, 0, 1, 0, 0, 0, 0, 0, 1, 1, 0, 0], "listed": true, "loved": [], "caveat": "", "short_quote": "Take a look at Karpathy's micrograd repo to become more familiar with the process.", "kind": "", "icon": false, "domain": null, "github_repo": null, "official_url": "https://github.com/karpathy/micrograd", "pricing": "", "aliases": ["micrograd"], "verdict": null, "praise": [{"label": "", "summary": "", "quotes": [{"id": 788220624, "item_id": 49263789, "url": "https://news.ycombinator.com/item?id=49263789", "by": "akhalilli", "date": "2026-08-11", "stance": "recommend", "quote": "if you want to build autograd instead of reading about it, Karpathy's micrograd / zero to hero", "excerpt": "author here. honest answer: there isn't much that covers the whole thing end to end, that's partly why I started writing this. but the ones I learned from: - Edward Yang's \"pytorch internals\" post: http://blog.ezyang.com/2019/05/pytorch-internals/ it's the classic, goes deep on the c++ side. - the pytorch developer podcast, same author. short episodes, one internal topic each - the pytorch repo's own CONTRIBUTING.md,", "context": "", "use_case": "Learning autograd by building", "switched_from": "", "story": "The whole of PyTorch on one page"}, {"id": 785388976, "item_id": 49086811, "url": "https://news.ycombinator.com/item?id=49086811", "by": "pyentropy", "date": "2026-07-28", "stance": "recommend", "quote": "Take a look at Karpathy's micrograd repo to become more familiar with the process.", "excerpt": "Inference (the real time computation) and training (the one you do when developing a model) are deeply tied by automatic differentiation. In a way, you train by repeatedly inferring, calculating a cost/loss (how much you suck) and then improving the model. Take a look at Karpathy's micrograd repo to become more familiar with the process.", "context": "", "use_case": "learning automatic differentiation", "switched_from": "", "story": "You Could Have Come Up with Kimi Delta Attention"}, {"id": 743751504, "item_id": 46484469, "url": "https://news.ycombinator.com/item?id=46484469", "by": "jerkstate", "date": "2026-01-03", "stance": "recommend", "quote": "Karpathy\u2019s micrograd did it first (and better)", "excerpt": "Karpathy\u2019s micrograd did it first (and better); start here: https://karpathy.ai/zero-to-hero.html", "context": "", "use_case": "Autograd teaching library", "switched_from": "", "story": "MyTorch \u2013 Minimalist autograd in 450 lines of Python"}]}], "caveats": [], "who": [], "tags": [], "use_cases": ["learning autograd by building", "learning automatic differentiation", "autograd teaching library"], "compared": [], "makers": [], "threads": [{"url": "https://news.ycombinator.com/item?id=49262961", "title": "The whole of PyTorch on one page", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=49085909", "title": "You Could Have Come Up with Kimi Delta Attention", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=48548904", "title": "DeepFork \u2013 reverse-engineer any OSS repo into a clean-room rebuild blueprint", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46483776", "title": "MyTorch \u2013 Minimalist autograd in 450 lines of Python", "mentions": 1}, {"url": "https://news.ycombinator.com/item?id=46454587", "title": "Build a Deep Learning Library", "mentions": 1}], "sentiment_by_month": {"2025-11": {}, "2025-12": {}, "2026-01": {"compare": 1, "recommend": 1}, "2026-02": {}, "2026-03": {}, "2026-04": {}, "2026-05": {}, "2026-06": {"neutral": 1}, "2026-07": {"recommend": 1}, "2026-08": {"recommend": 1}, "2026-09": {}, "2026-10": {}}, "quotes": [{"id": 788220624, "item_id": 49263789, "url": "https://news.ycombinator.com/item?id=49263789", "by": "akhalilli", "date": "2026-08-11", "stance": "recommend", "quote": "if you want to build autograd instead of reading about it, Karpathy's micrograd / zero to hero", "excerpt": "author here. honest answer: there isn't much that covers the whole thing end to end, that's partly why I started writing this. but the ones I learned from: - Edward Yang's \"pytorch internals\" post: http://blog.ezyang.com/2019/05/pytorch-internals/ it's the classic, goes deep on the c++ side. - the pytorch developer podcast, same author. short episodes, one internal topic each - the pytorch repo's own CONTRIBUTING.md,", "context": "", "use_case": "Learning autograd by building", "switched_from": "", "story": "The whole of PyTorch on one page"}, {"id": 785388976, "item_id": 49086811, "url": "https://news.ycombinator.com/item?id=49086811", "by": "pyentropy", "date": "2026-07-28", "stance": "recommend", "quote": "Take a look at Karpathy's micrograd repo to become more familiar with the process.", "excerpt": "Inference (the real time computation) and training (the one you do when developing a model) are deeply tied by automatic differentiation. In a way, you train by repeatedly inferring, calculating a cost/loss (how much you suck) and then improving the model. Take a look at Karpathy's micrograd repo to become more familiar with the process.", "context": "", "use_case": "learning automatic differentiation", "switched_from": "", "story": "You Could Have Come Up with Kimi Delta Attention"}, {"id": 776931297, "item_id": 48558206, "url": "https://news.ycombinator.com/item?id=48558206", "by": "gerardordz96", "date": "2026-06-16", "stance": "neutral", "quote": "The only worked example so far is karpathy/micrograd in the repo", "excerpt": "Ha, thank you. Fair nudge, it wasn't on any of the community lists yet, so I just opened a PR to add it to awesome-claude-skills: https://github.com/ComposioHQ/awesome-claude-skills/pull/1088 (awesome-claude-code is next). If you meant a showcase of what people have actually deepforked, that doesn't exist yet, honestly a good idea. The only worked example so far is karpathy/micrograd in the repo. What would you point", "context": "", "use_case": "", "switched_from": "", "story": "DeepFork \u2013 reverse-engineer any OSS repo into a clean-room rebuild blueprint"}, {"id": 743751504, "item_id": 46484469, "url": "https://news.ycombinator.com/item?id=46484469", "by": "jerkstate", "date": "2026-01-03", "stance": "recommend", "quote": "Karpathy\u2019s micrograd did it first (and better)", "excerpt": "Karpathy\u2019s micrograd did it first (and better); start here: https://karpathy.ai/zero-to-hero.html", "context": "", "use_case": "Autograd teaching library", "switched_from": "", "story": "MyTorch \u2013 Minimalist autograd in 450 lines of Python"}, {"id": 743320608, "item_id": 46457538, "url": "https://news.ycombinator.com/item?id=46457538", "by": "butanyways", "date": "2026-01-01", "stance": "compare", "quote": "\"micrograd\" is for scalars and we use tensors in the book", "excerpt": "If you are asking about the \"micrograd\" video then yes a little bit. \"micrograd\" is for scalars and we use tensors in the book. If you are reading the book I would recommend to first complete the series or atleast the \"micrograd\" video.", "context": "", "use_case": "", "switched_from": "", "story": "Build a Deep Learning Library"}]}