Agentic AI Engineering & Production Systems
Build autonomous tool-using AI agents, multi-modal LLM pipelines, LangChain security, and cloud sandbox isolation.
Course Overview & Objectives
Learn how to architect, build, and deploy production-grade autonomous AI agents equipped with tools, memory, and multi-agent coordination while enforcing military-grade security sandboxes around external code execution.
What You Will Master
- Build autonomous agents with function-calling, planning loops, and reflection capabilities
- Isolate code execution tools within Firecracker microVMs and secure gVisor sandboxes
- Audit LangChain and LlamaIndex agents for unintended tool-calling privilege escalation
- Implement multi-agent consensus protocols and rate limiting architectures
Prerequisites
- Intermediate Python
- Familiarity with REST APIs and async programming
Platforms & Tools Covered
Detailed Curriculum Modules
1 modules structured from foundational theory through complex adversarial execution.
Agent ReAct Loop & Function Calling Architectures
Reasoning, planning, and state management in autonomous LLM agents.
Hands-on Virtual Sandbox Labs
Zero local hardware dependencies. Provisioned in cloud containers via browser terminal.
Sandbox Isolation for Python Code Execution Agent
Deploy an agent that runs arbitrary code inside restricted gVisor microVM containers.
Faculty & Lead Instructor
Direct weekly instruction, live office hours, and code-review feedback.
Harpreet Kaur
Thread Security EducationSenior AI Systems Engineer
Building enterprise tool-using agent fleets and high-concurrency LLM inference microservices.
Frequently Asked Questions
Everything you need to know about scheduling, cohort admissions, and lab access.
Will I build real autonomous agents?
Yes! You build multi-agent workflows with LangGraph capable of automated research, coding, and triage.
Ready to Master Agentic AI Engineering & Production Systems?
Join the upcoming cohort. Seats are limited to maintain a high faculty-to-student ratio and rigorous sandbox feedback.