Small language models,
built from scratch.
Real SLMs, engineered end to end by our bootcamp cohort. Every showcase is a complete pipeline: pre-training, fine-tuning, quantization, GPU inference, and deployment to phone, laptop, and cloud.
1
Showcase
7
Pipeline stages
15-45
Min walkthroughs
What's inside every showcase
One link, the whole journey
A single website presenting every model they pre-trained and fine-tuned, and the process behind each one, systematic enough to teach a first-year undergrad.
Replicable by design
Each project ships an instruction.txt so you can reproduce the entire pipeline: data, training, quantization, and deployment.
Runs on your device
Offline downloads included: GGUF files, custom Android APKs, and desktop builds, with and without offline RAG.
A conference-style walkthrough
A 15-45 minute video where the builder walks through their website and explains what worked, what didn't, and why.
The showcase
1 engineer has shipped their SLM end to end. Order shuffles daily.
