AI Engineering Fundamentals

Descripción

Note: This course contains the use of AI. Stop watching AI tutorials. Start engineering AI systems; the scientific way. Most "Learn AI" courses teach you to copy a notebook and call it a day. This course is different. We treat AI engineering the way real scientists treat their work: with first principles, clean experiments, measurable outcomes, and honest reasoning about what works and why. Every module follows our Bricks → Walls → Castles model: small concepts → focused mini labs → full project builds. In this fundamental course of focused, project-based learning, you'll go from "I know Python" to "I can design, build, evaluate, and deploy real LLM applications." Every concept is taught before you use it. Every lab connects to the bigger picture. Every project ends with something you can put on your GitHub, your resume, and your portfolio. WHAT MAKES THIS COURSE DIFFERENT Scientist-led pedagogy. TechBricks is a team of scientists and engineers with 45+ years of combined experience across academia and industry. We teach intuition AND math; no black boxes, no hand-waving. Concept-first, then code. Every module follows our Bricks → Walls → Castles model: small concepts → focused mini-labs → full project builds. You'll never wonder "but WHY does this work?" Framework-light. You'll use the OpenAI SDK directly (no LangChain, no LlamaIndex, no magic). When you finish, you'll deeply understand what those frameworks do under the hood — and when NOT to use them. Real projects, not toy demos. You'll build a mini-Transformer from scratch, an LLM playground, a document data extractor, a multi-step reasoning engine, an injection-resistant chatbot, a chat-with-docs RAG system, a tool-using agent, an evaluation pipeline, and a production FastAPI service. Both cloud and local LLMs. Learn with OpenAI API AND with free local models via Ollama — so the course works whether you have a budget or not. PROJECTS YOU'LL ADD TO YOUR PORTFOLIO • Mini-Transformer (from scratch) • Interactive LLM Playground • Intelligent Document Data Extractor • Multi-Step Reasoning Engine • Injection-Resistant Secure Chatbot • Chat-with-Docs RAG Application • Tool-Using AI Agent (no frameworks) • Automated Evaluation Pipeline • Production-Grade LLM API with FastAPI 37 mini-labs + 13 project labs = 50 hands-on exercises. Enroll today and start building AI applications you actually understand. See you in Module 1. The TechBricks Team

Lo que aprenderás

• Build a mini Transformer from scratch and understand self-attention, multi-head attention, positional encoding, and GPT-style autoregressive generation
• Master tokenization (BPE, WordPiece, SentencePiece) and control LLM outputs using temperature, top-k, top-p sampling, streaming, and context windows
• Apply prompt engineering techniques: zero-shot, few-shot, chain-of-thought, ReAct, prompt chaining, and JSON-mode structured output extraction
• Defend LLM apps against prompt injection and jailbreaks by building a secure chatbot with input validation and defensive system prompts
• Set up a professional AI dev environment with API key hygiene, determinism controls, logging, and run open-source models locally using Ollama

Requisitos

• Basic Python knowledge. No prior AI, ML, or deep learning experience is required to get started.
• A computer with at least 8 GB of RAM and a stable internet connection.
• Curiosity and a willingness to think like a scientist: form hypotheses, run tests, and iterate based on evidence; we'll teach the rest.
• Familiarity with using a terminal or command line is helpful, but every command is walked through step by step.