TryCrucible — Prove your AI engineering skills

Scenario

You are given RAG questions, retrieved context, and generated answers. Build an eval suite that labels each answer as faithful, unsupported, contradicted, or partially supported, and explains the evidence behind the label.

Expected output

Eval suite correctly labels answer faithfulness and gives useful explanations for 24 held-out cases

Dataset

📦

Dataset files

Scoring rubric

Clarity5%

Eval report shows label, evidence, and rationale per answer

LLM usage20%

LLM-as-judge prompt requires quotes or evidence spans from context

Robustness10%

Handles partial support, subtle contradictions, and unsupported extra claims

Correctness25%

Labels match ground truth for at least 20/24 held-out cases

Architecture20%

Faithfulness, contradiction, and omission checks are separate and interpretable

Language-free evaluation

Build your solution in any language or framework — Python, TypeScript, Go, Rust, Java, C#, or anything else. The dataset artifacts may be in one language; your implementation does not need to match. TryCrucible evaluates the behaviour of your system, the quality of your AI workflow, your verification strategy, and the reproducibility of your submission — not your language choice.

Submission requirements

A public GitHub repository link
A Dockerfile in the repo root — any language or framework; the evaluator builds and runs your container
Your solution reads test_inputs.json from the working directory and writes results.json — standard I/O contract across all challenges
A decisions.md — 3–5 sentences on the key architectural and AI-workflow choices you made
The system must be fully reproducible — we clone, build, and run it against real test inputs

Evaluation contract

When you submit, the evaluator runs these steps in order:

1Clone your public GitHub repository
2Build your container from the Dockerfile in the repo root
3Mount test_inputs.json into the working directory
4Run your solution in a network-isolated sandbox (5 min limit, 512 MB RAM)
5Read results.json from the working directory
6Score correctness against hidden ground truth, then score architecture, AI workflow, robustness, and clarity

// I/O contract — same schema for every challenge

Input (provided by evaluator)

// test_inputs.json
[
  { "id": "t1", "input": { ... } },
  { "id": "t2", "input": { ... } }
]

Output (written by your solution)

Create a free account to start

Already have one? Sign in

Write evals for RAG answer faithfulness