Data Science Tutorials

Learn how to write a spec, turn it into a plan, and let Claude Code build using spec-driven development. Compare Superpowers, Spec Kit, and BMAD-METHOD to find the right tool for your workflow.

Connect Xiaomi MiMo-V2.5-Pro to OpenCode with Olostep MCP and the MiMo Token Plan, then put it to the test by building a CLI tool and a Reflex web app.

Singular Value Decomposition (SVD) is a matrix factorization method that breaks any matrix into three simpler components, revealing its underlying structure.

Run Qwen3.6 27B on an RTX 3090 and learn how Multi-Token Prediction (MTP) with llama.cpp can boost local LLM inference by almost 2x without upgrading your GPU.

Learn how OpenAI's gpt-realtime-2, gpt-realtime-translate, and gpt-realtime-whisper differ, then test each one with working Python WebSocket code.

Build a Python exercise trainer that logs workouts, remembers user history with Supermemory, and suggests personalized sessions across separate AI agent runs.

Learn what Perplexity Computer is, how its parallel sub-agents work, what credits really cost in a hands-on test, and where Pro and Max fit real research workflows.

Build an AI agent with Kimi K2.6, Olostep, and the OpenAI Agents SDK that reads your CV, finds live roles, filters noise, and tells you exactly where to apply.

Customize Claude Code in the terminal with permissions, CLAUDE.md, plan mode, /loop, hooks, /voice, and cost tracking. Seven upgrades that pay off fast.

Learn how to run the full DeepSeek V4 Flash model on a single GPU using a modified llama.cpp build and a compatible GGUF file in this hands-on tutorial.

A practical overview of contrastive learning - how models learn by comparing similar and dissimilar examples, the loss functions and methods behind it, and how to implement it in PyTorch.

A conceptual guide to the kernel trick - what it is, how it enables SVMs and other kernel-based models, and when to use it over other approaches to nonlinear modeling.