CourseAgentic AI Engineer: The Complete Course · Module 17: Professional Practice · part 96 of 97
Part 96 · Module 17: Professional Practice

Lab: Ship a repository you can show

17 min read·13 Oct 2026
Module 17 · Lab · About 60 minutes · Needs: lessons 1 and 2 of this module, and the Module 13 lab (evals/baseline.json and the gate)

What you are building

RECORDS IN, TWO DOCUMENTS OUT baseline.json shipped run evals/runs/ older runs evals/report.py checks every number docs/write-up.md for the review README.md for the public then a checklist of all 21 files the course built
One new file (blue: your system) turns saved run records into two checked documents (green: passed), so no number is typed by hand.

It is Thursday 6 August 2026, week 17, the morning of the one-month review (simulated). This lab turns the repository you have built since Module 1 into something you can show. One new file, evals/report.py, reads the run records your eval gate saves and generates the two files the course guide names for this module: docs/write-up.md for Priya and Marta, and README.md for anyone who opens the repository. The write-up carries Lesson 1's honest sentence, limits card and scope review; both files pass Lesson 2's claim check and README lint, and one command regenerates and checks them in CI. The lab ends with a checklist of every file the course built. It makes no model calls: every number comes from a saved record. Everything at BrightCart is simulated.

The rest of this course is yours to keep

This course is bought on its own, once, and stays readable afterwards, including the parts added to it later.

Buy this course · ₹399See what is inside