AI Engineering Primer | From Basics to Advanced | 2026

By Saurav Prateek

Master production-grade artificial intelligence engineering by building autonomous agents, advanced workflows, and intelligent RAG systems from scratch. Learn to leverage LangChain, LangGraph, and modern LLMs to architect scalable, tool-enabled AI applications capable of complex reasoning and self-reflection.

Table of Contents

AI Engineering Primer | From Basics to Advanced | 2026

Table of Contents

AI Agents and AI Workflows are not the same | Agents Vs Workflows | Code walkthrough | Python

Model Context Protocol (MCP) from scratch | Connecting your AI Agents with external server and tools

Tool Calling - Make LLMs talk to Database, APIs and Web | Langchain | Python

Human in the Loop | Interrupting your Agentic Workflow for user input | LangChain | Python

Optimize your Agentic workflow with Parallel Execution in LangGraph | LangChain, Python

Vector Embeddings for Semantic Analysis | Introducing Google Gemini Embedding Models

Add Memory to your AI Agents | Context Management for LLMs | LangChain

Debugging your Agentic Workflows | Threads & Checkpoints | LangChain

Building a Deep Research Agent - Static Workflow | Dissecting my Static-DRA Research Paper | Python

Google Self-Evolution algorithm for Deep Researcher | Agentic Workflow | Parallel Processing, Python

Agent Builder by OpenAI | Introduction to building Agentic Workflows | No-Code

Learn Agent2Agent (A2A) Protocol | Build Your First Multi-Agent Network | Python

Prompt Chaining | Agentic Workflow Design Pattern | LangChain, LangGraph | Python

Routing - Agentic Workflow Design Pattern | Agent summarizing social media profile | LangChain

Evaluator - Optimizer Design Pattern | Agentic Workflow | Human in the Loop, LangChain, Python

Orchestrator - Worker Design Pattern | Agentic Workflow | Powering Deep Research agents

Building a Self-RAG | Making your Agentic workflows critique through Self-Reflection | LangChain

Building a Corrective RAG | Integrating Web Search as a fallback knowledge base | LangChain | Python

Building Agentic workflow that creates personalized Linkedin Post | Few Shot Prompting | LangChain

Building a Referral Outreach Agentic Workflow in LangChain | LangGraph, Python (from scratch)

Building a Text to SQL Agent for handling complex database queries | LangChain | Python

Agent that recommend Roles at Google, reviews Resume and outlines Preparation strategies | LangChain

Understanding LoRA from scratch | Dissecting Low-Rank Adaptation of LLMs Research Paper

Implement LoRA in LLMs from scratch | Modifying Llama 3 architecture | PyTorch | Python

Introduction to Google FunctionGemma 270M model - LLM for Tool Calling

Evaluating LLM on HumanEval coding benchmark | Gemini 3.1 Pro | Google

Visualizing Byte-Pair encoding Tokenization process in LLM | HuggingFace | Python

Mixture of Experts (MoE) spelled out | Implement from scratch | Pytorch | Python

Reproduce LLM for Storytelling (3.6M) - TinyStories | HuggingFace | Python

Fine-tune GPT-2, 125M model for text classification | Pytorch | Python

Fine-Tune LLM on Text Summarization task from scratch | Pytorch | Python

Fine-Tuning Llama 3.2 - 1B model with LoRA from scratch | Text summarization | PyTorch | Python

Reproducing Google FunctionGemma 270M model | LLM for Tool Calling

Building a Character level Language Model from scratch | Neural Networks | PyTorch | NLP

Building a Language Model for Auto-completion and Suggestions | Neural Networks | PyTorch | NLP