Build a tested AI agent in Python—not another chatbot demo. Across this free 10-part course, you’ll build IST Travels with tools, memory, specialist agents, guardrails, evaluations, safe API integration, and FastAPI.
VERIFIED COURSE EVIDENCE
✅ 63 automated tests pass
✅ 25/25 acceptance gates pass
✅ 7/7 behavior-evaluation baseline
✅ 11-span local execution trace
✅ Intentional regression detected
✅ Real local FastAPI HTTP smoke test
✅ Core path works without a paid API
By the end, you will be able to explain, build, test, trace, and defend an agent system—not merely copy prompts.
This is Step 1. You’ll understand the controlled agent loop, see how an agent differs from a chatbot or fixed workflow, and learn what the model may judge versus what the application must control.
This course is for Python developers, solution architects, technical product builders, and AI-curious engineers who want practical patterns instead of disconnected demos.
IST Travels uses reproducible sample providers in the core learning path. It plans and compares; it does not book travel or claim live prices or availability.
Subscribe to complete all 10 lessons. New steps release Tuesday, Thursday, and Saturday at 6:00 PM ET through September 29.
Full course: https://www.youtube.com/playlist?list=PLQA5J_JhHm7I
Next lesson: https://youtu.be/qhMjCa5qSo0
Chapters:
00:00 Learn agents by building one
00:43 Concrete first, mechanism next, evidence always
01:18 What you need before starting
01:56 Model judgment inside a controlled loop
02:45 Agents vs chatbots and workflows
03:31 From answers to bounded work
04:23 Grow one agent into a tested system
05:07 Watch, build, break, and prove
05:48 Run the course without a paid API
06:27 What you will be able to build
#AIAgents #Python #AgenticAI
VERIFIED COURSE EVIDENCE
✅ 63 automated tests pass
✅ 25/25 acceptance gates pass
✅ 7/7 behavior-evaluation baseline
✅ 11-span local execution trace
✅ Intentional regression detected
✅ Real local FastAPI HTTP smoke test
✅ Core path works without a paid API
By the end, you will be able to explain, build, test, trace, and defend an agent system—not merely copy prompts.
This is Step 1. You’ll understand the controlled agent loop, see how an agent differs from a chatbot or fixed workflow, and learn what the model may judge versus what the application must control.
This course is for Python developers, solution architects, technical product builders, and AI-curious engineers who want practical patterns instead of disconnected demos.
IST Travels uses reproducible sample providers in the core learning path. It plans and compares; it does not book travel or claim live prices or availability.
Subscribe to complete all 10 lessons. New steps release Tuesday, Thursday, and Saturday at 6:00 PM ET through September 29.
Full course: https://www.youtube.com/playlist?list=PLQA5J_JhHm7I
Next lesson: https://youtu.be/qhMjCa5qSo0
Chapters:
00:00 Learn agents by building one
00:43 Concrete first, mechanism next, evidence always
01:18 What you need before starting
01:56 Model judgment inside a controlled loop
02:45 Agents vs chatbots and workflows
03:31 From answers to bounded work
04:23 Grow one agent into a tested system
05:07 Watch, build, break, and prove
05:48 Run the course without a paid API
06:27 What you will be able to build
#AIAgents #Python #AgenticAI
- Категория
- Бесплатные курсы
Комментариев нет.
