The ML API for engineers who don't have time for ML
Hey HN - spending time adding machine learning to your product is almost never worth it for startups. Formulating the problem, managing training data, training models, serving them, etc... for a 10% increase in some metric just isn't worth the time. Using rules + heuristics, although not optimal, does a pretty good job for a long time. There should be a better solution between rules & hardcore ML -- an 80/20 way to add ML to any product. Just slap it on top of the rules you already have, get that 10% gain, and never think about it again. I built an API that does just that. It's used by…
What it does
In the maker’s words, at launch
Hey HN - spending time adding machine learning to your product is almost never worth it for startups. Formulating the problem, managing training data, training models, serving them, etc... for a 10% increase in some metric just isn't worth the time. Using rules + heuristics, although not optimal, does a pretty good job for a long time. There should be a better solution between rules & hardcore ML -- an 80/20 way to add ML to any product. Just slap it on top of the rules you already have, get that 10% gain, and never think about it again. I built an API that does just that. It's used by a few companies to do everything from shopping cart recommendations to personalized landing page hero text. It takes 10-20 lines of code to implement. I'd love for anyone in the HN community to kick the tires on this and let me know what you think. Docs are at https://www.banditml.com/. You can even use this api key immediately without an account to test it out: 1e9d8ef7948e545d25f189addb17dae1.
Does the same job
all alternatives →
- IMI'm 15 and built a neural network from scratch in C++,just math2025 · github.com · ▲8
I’m 15 and self-taught. I'm learning ML from scratch because I want to really understand how things work. I’m not into frameworks. I prefer math, logic, and C++. I implemented a basic MLP that supports different activation and loss functions. It was trained via mini-batch gradient descent. I wrote it from scratch, using no external libraries except Eigen (for linear algebra). I learned how a Neural Network learns (all the math) -- how the forward pass works, and how learning via backpropagation works. How to convert all that math into code. I’ll write a blog soon explaining how MLPs work in…
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 27d ago · cactuscompute.com
Launched alongside, August 2021
the whole month →- II
Hey all My name is Sindre, and I am the CTO of Scrimba (YC S20). For the last 7 years, I have written all my web apps in a full-stack programming language called Imba. It compiles to JavaScript and its main goal is to make web developers more productive. I just launched a major overhaul of Imba, so I wanted to share it here on HN, in case anyone are interested in learning more about it. It is very opinionated, so some of you might not like it, but I would love to hear anyones feedback regardless. Constructive criticism appreciated! The backstory: Imba initially started in 2012 as an effort…
Life & fun · 2021
- WB
Hello HN, Over the last year we've been building ente[1], a privacy-friendly, easy-to-use alternative to Google Photos. We've so far built Android[2][3], iOS[4], web[5] apps that encrypt your files and back them up in the background. You can access these across your devices, and share them with other ente users, end-to-end encrypted. You can also use our electron app[6] to maintain a local copy of your backed up files. We've built a fault-tolerant data replication layer that replicates your data to two different storage providers in the EU. We will be providing additional replicas as an…
Life & fun · 2021
