Data Science Tutorials

Learn how to build a custom AI agent harness with Python, ReAct loops, and durable state.

Learn how to use Anthropic’s latest flagship model to build a Python agent that reads a Flask repository before planning a change. Add progress updates, read-only file tools, and cost controls.

Use Unsloth Desktop for local AI workflows, including LLM inference, fine-tuning, vision models, image and video generation, OpenCode coding agents, and local APIs from one desktop application.

Learn how quantization, KV-cache optimization, batching, speculative decoding, memory bandwidth, and hardware-aware tuning can improve speed, scalability, and cost efficiency when serving large language models.

Use GLM-5.3-Flash and CrewAI to build 3 specialist LLM agents covering stock fundamentals, technicals, and news, and merge their findings into a BUY/SELL signal.

Learn how to run Qwen3.8-Flash-Next GGUF locally with llama.cpp on an RTX PRO 6000, then connect it to OpenCode for a fully local agentic coding setup.

Learn how Grok Bot's Bots, shared cloud computer, skills, and routines fit together by building one agent that researches, verifies, and reports every Monday.

Learn how to run Qwen3.8-27B locally with Blackwell-native NVFP4 and MTP speculative decoding, achieving up to 170 tokens per second with llama.cpp.

Learn how to harness SpaceXAI’s newest frontier model to build intelligent, tool-using agents from scratch. This comprehensive guide walks you through everything from basic API setup to deploying a fully autonomous loop with prompt caching and custom tools.

Learn how Grok Build and Claude Code differ in practice: I planted three defects in a dataset and ran the same five-turn conversation through both agents.

Learn how to set up and test DeepSeek Harness, an open-source AI coding agent with plugin support, web search, vision, DeepSeek models, Gemini, ModLens, third-party models, and agentic workflows.

A hands-on look at Anthropic's Fable 5 model in Claude Code: how it handles a one-shot build, its safety refusals and Opus fallback, and when to use it.