Google Data Analytics, IBM AI & Meta Marketing — All in One Subscription
Most AI Pilots Fail to Scale. MIT Sloan Teaches You Why — and How to Fix It
Overview
Google, IBM & Meta Certificates — All 10,000+ Courses at 40% Off
One annual plan covers every course and certificate on Coursera. 40% off for a limited time.
Get Full Access
Learn to build and deploy AI agents from scratch through a comprehensive tutorial covering the complete development lifecycle from basic implementation to production deployment. Start by understanding what AI agents are and their fundamental concepts, then progress through hands-on development using LangChain and Ollama to create a simple local AI agent. Expand your agent's capabilities by integrating single and multiple tools while implementing streaming functionality and conversation history management. Master observability by implementing tracing for AI agent calls using MLflow to monitor and debug agent behavior. Transform your agent into a production-ready application by building a REST API with FastAPI, containerizing the solution with Docker and Docker Compose, and finally deploying to a production environment using Render. Gain practical experience with modern AI development tools and deployment strategies while building a fully functional AI agent system ready for real-world applications.
Syllabus
00:00 - Welcome
00:53 - What is an AI agent?
03:34 - Project setup & dependencies
04:21 - Simple local AI agent - Langchain & Ollama
08:57 - Single tool agent - tools, streaming and conversation history
13:37 - Multi-tool agent - give more tools to your agent
15:00 - Tracing AI agents calls with MLflow
17:15 - RestAPI with FastAPI
21:10 - Dockerfile - container definition and Docker compose
24:37 - Deploy to production Render setup
27:30 - Conclusion
Taught by
Venelin Valkov