DS410BKK
Generative AI in product development, principles, techniques and technology behind it
Faculty
Maxim Musin
CEO at rebels.ai
Course length
Duration
Total hours
Credits
Language
Course type
Fee for single course
Fee for degree students
Skills you’ll learn
Overview
This intensive 15-day course provides comprehensive training in building production-ready products using state-of-the-art agentic AI techniques. Students will master IDE-integrated agentic tools (Windsurf, Antigravity, Claude Code), explore the emerging “vibe coding” paradigm, implement agent communication protocols (MCP and A2A), and design fault-tolerant, cost-optimised multi-agent systems.
The curriculum addresses critical production challenges, including bounded autonomy, governance, FinOps optimisation, and enterprise-scale deployment using Amazon Bedrock AgentCore. By the end of the course, students will have hands-on experience across the full agentic development lifecycle — from intent-driven coding to production deployment — culminating in a capstone project that demonstrates their ability to build intelligent, autonomous systems.
Learning highlights
- IDE Agentic Tools Mastery: Work with Windsurf Cascade, Google Antigravity, and Claude Code for autonomous development workflows.
- Vibe Coding Paradigm: Master intent-driven development where natural language prompts generate production code.
- Protocol Integration: Implement both MCP (Model Context Protocol) and A2A (Agent-to-Agent) protocols for comprehensive agent communication.
- Agent Orchestration: Build multi-agent systems using LangGraph, CrewAI, and OpenAI Agents SDK with durable execution patterns.
- Production Readiness: Address the reality that
- Governance & Security: Design bounded autonomy architectures with governance agents, escalation paths, and compliance frameworks.
- Cost Optimisation: Achieve 90% cost reduction through Plan-and-Execute patterns, heterogeneous models, and strategic caching.
- Enterprise Deployment: Deploy to Amazon Bedrock AgentCore with workflow redesign and change management strategies.
- Capstone Project: Build and deploy a complete agentic product from specification to production.
Course outline
15 classes
Session 1
Introduction to IDE-integrated agentic development.
Hands-on with Windsurf Cascade and Google Antigravity.
Comparative analysis and tool selection strategies.
Session 2
The vibe coding paradigm shifts Intent-driven development workflows.
Prompt engineering for production code generation.
Session 3
State-of-the-art LLM landscape.
Model selection and context window strategies.
Hands-on model comparison and benchmarking.
Session 4
Model Context Protocol architecture and design.
Building MCP servers with Python.
SDK Security, access control, and advanced patterns.
Session 5
Agent-to-Agent Protocol fundamentals.
Cross-platform agent communication.
Multi-protocol integration (MCP + A2A).
HOMEWORK DUE: Multi-Protocol Agent System.
Session 6
Product analysis methodologies.
Project ideation and scope definition.
Architecture planning.
PROJECT KICKOFF: Students select capstone projects.
Session 7
E2E testing with Playwright.
AI-generated tests and automation.
Visual regression testing for agentic apps.
Session 8
Agentic planning and autonomous workflows.
Test-driven development for agent systems.
Debugging and code review with AI assistance.
Session 9
Orchestration patterns and frameworks. generation (e.g., RunwayML).
Multi-agent systems with LangGraph and CrewAI.
Swarm patterns and role-based agents.
Session 10
State persistence in agent workflows.
Time-travel debugging with LangGraph 1.0.8.
Building fault-tolerant agentic systems.
Session 11
Bounded autonomy architecture.
Governance agents and escalation paths.
Security, compliance, and audit trails.
Session 12
Cost-performance analysis for agentic systems.
Heterogeneous model strategies (90% cost reduction).
Plan-and-Execute pattern and caching techniques.
HOMEWORK DUE: Governance & FinOps Framework.
Session 13
RAG fundamentals and architecture.
Vector databases and hybrid search.
Advanced patterns for project knowledge systems.
Session 14
Perplexity AI integration for research.
Research workflow automation.
Analytics-driven development practices.
Session 15
Enterprise deployment patterns (addressing less than 25% success rate) Amazon Bedrock AgentCore deep dive CI/CD pipelines and push-to-deploy workflows.
FINAL PRESENTATIONS: Project demos.
Prerequisites
Python - Advanced Level (async/await, type hints, decorators)
AI/ML Fundamentals - Intermediate Level (LLMs, embeddings, vector databases)
API Development - Intermediate Level (REST, async APIs)
Cloud Services - Basic Level (AWS/GCP/Azure fundamentals)
Git/Version Control - Intermediate Level
Methodology
We will study a set of practical Jupyter notebooks, interrupted by relatively short theoretical parts. There will be two big homework assignments designed to emulate a relatively real data science project. There will also be personal projects based on Python integrations and capabilities of data analysis; this will be a good example of time management in a DS project. Finally, students will have a final exam and a student project demonstration at the end of the course.
Grading
Maxim Musin comes from a background in statistics, advanced multidimensional probability, and random processes. During his career in these fields, he found himself developing skills and gathering experience through working in both academic environments and the private sector. For the last 5 years Maxim is a CEO of for profit AI development laboratory rebels.ai, integrating AI in enterprise and helping startups reach the orbit.
His academic experience ranges from teaching probability and statistics at MSU and MIPT, as a member of the faculty of innovation and high technology, FIHT, which at the time was among the few places worldwide with capabilities for advanced statistics study. During his time there, he produced several notable projects with his students, particularly in regards to the stochastic convergence of neural networks. His course on applied modern statistics became mandatory for the data analysis division of the FIHT MIPT Masters.
See full profileApply for this course
Generative AI in product development, principles, techniques and technology behind it
by Maxim Musin
Total hours
45 Hours
Dates
Mar 16 - Apr 03, 2026
Fee for single course
€1500
Fee for degree students
€750
How to secure your spot
Complete the form below to kickstart your application
Schedule your Harbour.Space interview
If successful, get ready to join us on campus
FAQ
Will I receive a certificate after completion?
Yes. Upon completion of the course, you will receive a certificate signed by the director of the program your course belonged to.
Do I need a visa?
This depends on your case. Please check with the Spanish or Thai consulate in your country of residence about visa requirements. We will do our part to provide you with the necessary documents, such as the Certificate of Enrollment.
Can I get a discount?
Yes. The easiest way to enroll in a course at a discounted price is to register for multiple courses. Registering for multiple courses will reduce the cost per individual course. Please ask the Admissions Office for more information about the other kinds of discounts we offer and what you can do to receive one.