★ Your Lab Notebook - Validation Reports

Get the code to run this lab

The commands on this page come from the repository, not the website. Clone it and enter this lesson's folder: git clone https://github.com/ld-singh/ai-factory-ops-lab && cd ai-factory-ops-lab/portfolio-lab/06-validation-reports. Browse this lesson on GitHub

Course home: AI Factory Operations Lab

This directory is where claims become concrete. A lesson is only "Complete" when its report here holds real, captured output. No evidence directory, no completion - that rule is what separates "I read about it" from "I ran it and watched it happen."

How to use it

Each time you finish a runnable lesson:

1. Run the lesson's make ...-evidence (or collect-*-evidence.sh) step. Snapshots land in evidence/ as timestamped directories.
2. Open the matching report below and fill in your environment details plus a reference to that evidence directory.
3. Only then mark the lesson done.

The reports

Report	Lesson	Mode	Fill in after…
local-simulation-report.md	1 - K8s GPU scheduling	🟦 Sim	make phase1-up/-demo/-evidence
slurm-gres-validation.md	2 - Slurm	🟦 Sim (+ real GRES in Lesson 6)	Phase 3
real-gpu-validation-report.md	6 - Real GPU (Part A)	🟥 Real	The Lesson 6 hardware run ✅
hami-isolation-validation.md	6 - Real GPU (Part B - HAMi)	🟥 Real	The Lesson 6 Part B run ✅
inference-benchmark-report.md	6 - Real GPU (Part C - inference)	🟥 Real	✅ Validated — RTX A6000, 7B (concurrency sweep + saturation knee)

The limitations ledger - read this to grade your own claims

fake-vs-real-limitations.md is the single source of truth for what each lab mode can and cannot prove. Before you write any claim in a report, check it against that ledger.

➡️ Back to: the Learning Path.