Video
Design, build, and deploy intelligent AI agents.
Follow the same graduate-level curriculum taught at Stanford, adapted for working professionals.
This program guides you through end-to-end agent workflows, covering the latest techniques in self-improving AI agent tool use, reasoning, and planning. You’ll go beyond basic LLM prompting to develop agents that continuously improve through interaction with their environment. Drawing on Stanford research and recent advances in the field, you’ll learn both the theoretical foundations and practical engineering skills needed to design, evaluate, and deploy robust AI agents for real-world applications.
- Apply test-time scaling
- Explore methods for self-improvement and tool use
- Add retrieval and long-term memory to LLMs
- Build multi-step reasoning capabilities
- Design robust evaluation frameworks
Here's what you can expect...
Stanford Professors
Learn from Stanford Adjunct Professor Chowdhery and Assistant Professor Mirhoseini.
Hands-On Assignments
Complete coding and system-building assignments. Build a small-scale AI agent.
100% Online
Watch on-demand video lectures, attend live virtual sessions, and share in a peer discussion forum.
Earn a Certificate
Earn a Stanford Certificate of Completion in Agentic AI.
Who should enroll?
AI & ML Engineers
Machine Learning Engineers looking to build and improve AI agents, AI Engineers / AI Practitioners building agent-based systems, and AI Researchers and technical practitioners exploring agentic systems
Software Engineers & Developers
Software Engineers and technical professionals developing AI-powered applications and developers working with LLM APIs who want to build more advanced workflows
AI Practitioners
AI practitioners improving existing LLM applications using newer research ideas
Decision Makers Advancing Existing Systems
CTOs/VPs of Engineering at AI-forward companies who need to understand the technical depth of modern agentic systems
Lorem Ipsum Dolor sit Amet
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Donec nunc purus, mattis rutrum est ut, finibus fringilla mauris. Mauris arcu massa, hendrerit et lobortis eu, scelerisque a felis. Curabitur sed magna purus. Lorem ipsum nunc purus.
Graduate or Professional?
Time Commitment
Achievement
Classmate Interactions
Cost
Graduate Certificate Courses
90–120 hours per course
Earn up to 18 units of academic credit that may contribute to a certificate or a degree
Frequent collaboration with other students taking courses at the same time
$$$$
Professional Certificate Courses
6–13 hours per course
Earn a certificate and, in some cases, professional education units
Potential to connect with other participants through private social media groups
$$
Interested in Artificial Intelligence? Check out these online certificate programs:
Lorem ipsum dolor
Venenatis laoreet lacus
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nullam varius mi vitae ornare dapibus. Fusce risus mauris, convallis in turpis id, venenatis laoreet lacus.
Lorem ipsum dolor
Venenatis laoreet lacus
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nullam varius mi vitae ornare dapibus. Fusce risus mauris, convallis in turpis id, venenatis laoreet lacus.
Lorem ipsum dolor
Venenatis laoreet lacus
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nullam varius mi vitae ornare dapibus. Fusce risus mauris, convallis in turpis id, venenatis laoreet lacus.
Lorem ipsum dolor
Venenatis laoreet lacus
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nullam varius mi vitae ornare dapibus. Fusce risus mauris, convallis in turpis id, venenatis laoreet lacus.
Lorem ipsum dolor
Venenatis laoreet lacus
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nullam varius mi vitae ornare dapibus. Fusce risus mauris, convallis in turpis id, venenatis laoreet lacus.
Lorem ipsum dolor
Venenatis laoreet lacus
Lorem ipsum dolor sit amet, consectetur adipiscing elit. Nullam varius mi vitae ornare dapibus. Fusce risus mauris, convallis in turpis id, venenatis laoreet lacus.
Lorem ipsum dolor
Venenatis laoreet lacus
- list item 1
- list item 2
- list item 3
Lorem ipsum dolor
Venenatis laoreet lacus
- list item 1
- list item 2
- list item 3
Lorem ipsum dolor
Venenatis laoreet lacus
- list item 1
- list item 2
- list item 3
Lorem ipsum dolor
Venenatis laoreet lacus
- list item 1
- list item 2
- list item 3
Lorem ipsum dolor
Venenatis laoreet lacus
- list item 1
- list item 2
- list item 3
Instructors
Aakanksha Chowdhery
Adjunct Professor, Stanford Computer Science
Azalia Mirhoseuini
Assistant Professor, Stanford Computer Science
Curriculum Outline
Test-Time Scaling
Explore test-time scaling: using extra compute during inference to boost model performance. You will examine methods for generating and selecting outputs, and how to balance compute cost against quality.
Self-Improvement: Rewards & Verification
Examine verification and how output and process rewards differ and affect training and inference performance. You will explore reward model design via the Math-Shepherd paper, where verifier ensembles can outperform a single model.
Self-Improvement Techniques
Study the ReAct method, where agents pair reasoning with feedback to solve complex tasks grounded in real-world information. You will explore reinforcement learning from execution feedback and verification signals, and how models learn from their outputs.
Reasoning and Self-Improving Agents
Examine how AI agents plan and perform reasoning through tree search or parallel reasoning, and RL training with synthetic data for tool use. You will explore how LLMs support discovery, from ideation to paper writing.
Agentic Frameworks for SW Engineering & Augmenting LLMs with Memory
Explore how AI agents support software development, including CUDA kernel development and compiler optimization, an area where agents have advanced fastest. You will examine extending agent memory using ideas from MemGPT and Cartridges.
Long Horizon Tasks and Agentic Evaluations
Examine how to measure what tasks agents can complete as complexity and duration increase, from minutes-long to hours-long work and well-specified versus judgment-heavy problems. You will explore where agents struggle, like sourcing information, and how economic value reveals gaps between what agents attempt and finish.
Next Frontiers in Agentic AI
Recap the course, reviewing four papers on self-improving agents. You will close with future research directions and efficiency gains as inference costs rise.
Prerequisites
- Python programming experience
- Basic familiarity with LLMs, including how to use LLM APIs
- Basic understanding of machine learning concepts
