█
LastWrite
  • > Curriculum
  • > Pricing
  • > For Educators
  • > About
  • > Contact
Log InGet Started

Questions, concerns, bug reports, or suggestions? We read every message, write to us at [email protected].

More ways to reach us →
LastWrite

Structured computer science lessons for aspiring developers and security professionals.

[email protected]

(201) 785-7951

Mon–Fri, 9 AM–5 PM EST

Learn

  • Curriculum
  • Pricing

Company

  • About
  • For Educators & Schools
  • Contact Us

Legal

  • Terms of Service
  • Privacy Policy
© 2026 LastWrite. All rights reserved.
Curriculum/LLM Research and NLP/Evaluation and Experimental Rigor/Passion Project: Evaluation Harness
240 minAdvanced

Passion Project: Evaluation Harness

After this lesson, you will be able to: Build a reusable evaluation harness for a task that computes the right metrics with confidence intervals and significance tests, and apply it to compare two systems.

The Evaluation capstone: a harness you can point at any two systems and get a trustworthy verdict. This is the tool that makes every later claim in your research defensible, and reviewers love a paper whose evaluation is clearly rigorous.

Prerequisites:Reproducibility

The build

Pick a task and build an evaluation harness that: loads a (decontaminated, documented) test set, runs one or more systems with a fixed prompt/template, computes appropriate metrics (an overlap metric plus an embedding or judge metric where relevant), reports per-example scores, and produces mean plus-or-minus variance across seeds with a paired significance test and confidence intervals between systems. Make it config-driven and seeded so a run is one command and fully reproducible.

Milestones

A clean, reusable tool, not a one-off notebook.

  1. 1

    Milestone 1: Choose the task + metrics; justify why those metrics fit (and their limits). Document the test set + contamination check.

  2. 2

    Milestone 2: System runner: given a model + prompt template, produce predictions for the test set (fixed, logged).

  3. 3

    Milestone 3: Metric layer: compute per-example scores for each metric via the evaluate library; store them.

  4. 4

    Milestone 4: Statistics layer: mean + std across seeds, paired bootstrap p-value + 95% CI between two systems.

  5. 5

    Milestone 5: (If applicable) an LLM-as-judge mode with order-swapping + a small human-validation subset.

  6. 6

    Milestone 6: A report it generates automatically (table with means, CIs, significance) + a README and pinned environment so anyone can rerun it.

💡 Why this is the highest-leverage artifact

Most student papers are rejected on evaluation, so a reusable, rigorous harness is the thing that most raises the quality of everything you do next. It turns your fine-tuning project's 'it seems better' into 'it is better by X, 95% CI [..], p < 0.05, three seeds, on a decontaminated set.' That sentence is the difference between an anecdote and a result, and this harness produces it on demand.

How to talk about it

'I built an evaluation harness that runs systems on a decontaminated test set, computes overlap and embedding metrics plus an order-swapped LLM judge, and reports significance with a paired bootstrap across seeds.' That tells an advisor you understand the part of research that actually gates publication. Use it on your own fine-tuning project to produce the headline comparison, and reuse it for the reproduction in the Publishing sub-track.

Sign in and purchase access to unlock this lesson.

Sign in to purchase
←Reproducibility
Back to Evaluation and Experimental Rigor
Evaluation Job Readiness→