Home/Deals/IT & Software

Master LLM Engineering & AI Agents: Build 14 Projects - 2025

Learn Hugging Face, LangGraph, CrewAI, AutoGen, N8N, RAG, MCP, & OpenAI Agents SDK + Expert Help & Access AI Community

4.7 ★★★★★5,210 studentsCreated by Prof. Ryan Ahmed, Stemplicity Inc.Last updated Oct 14, 2025🌐 English

What you'll learn

✓Understand the foundations of Large Language Models (LLMs) and Agentic AI, including how LLMs are trained, fine-tuned, and deployed.

✓Create and deploy intelligent autonomous AI agents using cutting-edge frameworks like AutoGen, OpenAI Agents SDK, LangGraph, n8n, and MCP.

✓Explore and benchmark open-source LLMs such as LLama, DeepSeek, Qwen, Phi, and Gemma using Hugging Face and LM Studio.

✓Develop real-world applications using API access to OpenAI, Gemini, and Claude for text generation and vision tasks.

✓Apply a proven 5-step framework to select the right AI model for your business: maximizing cost-efficiency, minimizing latency, & accelerating time to market.

✓Evaluate LLMs using leaderboards like Vellum and Chat Arena, and conduct blind tests to objectively assess AI model performance.

✓Design Retrieval-Augmented Generation (RAG) pipelines using LangChain, OpenAI embeddings, & ChromaDB for efficient document retrieval & question answering.

✓Build an interactive, transparent AI-powered Q&A system with a Gradio interface that displays answers along with source citations for enhanced user trust.

✓Master data validation & structured output generation using the Pydantic library, including BaseModel, type hints, & parsed output creation from OpenAI models.

✓Build an AI-powered resume editor that analyzes gaps between a resume & job description & automatically tailors resumes/cover letters for targeted applications.

✓Learn how to fine-tune pre-trained open-source LLMs using parameter-efficient methods like LoRA and tools such as Hugging Face’s TRL and SFTTrainer.

✓Master dataset preparation and model evaluation techniques, including calculating accuracy, precision, recall, and F1-score using scikit-learn.

✓Apply key components in Hugging Face Transformers library such as pipeline( ), AutoTokenizer( ), and AutoModelForCausalLM( ).

✓Gain practical experience working with open-source datasets/models on Hugging Face, & apply quantization techniques like bitsandbytes to optimize Performance.

✓Master advanced prompt engineering techniques such as zero-shot, few-shot, and chain-of-thought prompting.

✓Deploy multi-model AI agents using AutoGen, integrating LLMs from OpenAI, Gemini, & Claude, enabling agent collaboration & human-in-the-loop oversight.

✓Develop and deploy agentic AI workflows using LangGraph, mastering concepts like states, edges, conditional logic, and multi-stage nodes.

✓Design & build AI-powered booking agents using LangGraph, enabling automated search & recommendation of flights & hotels through integration with external APIs.

✓Build a data science agent team using CrewAI, creating specialized agents for workflow planning, data analysis, model building, and predictive analytics.

✓Design and automate end-to-end Agentic AI workflows using n8n, integrating services like Gmail, Google Sheets, Google Calendar, and OpenAI.

✓Build an advanced AI tutor system using Model-Context-Protocol (MCP) and OpenAI Agents SDK, enabling dynamic tool interoperability.

✓Apply classical ML models (linear regression, random forest, XGBoost) within agent workflows, including dataset loading and inspection.