TryCrucible — Prove your AI engineering skills

Scenario

AcmeWorks has 160 internal policy documents covering employee and technical workflows. Some questions require combining facts across sections, regions, and approval rules. Build a retrieval and answer-generation pipeline that cites the right source documents and avoids single-document shortcuts.

Expected output

Answers combine all required facts and cite the relevant source document IDs for 12 multi-hop questions

Dataset

📦

Dataset files

Scoring rubric

Clarity5%

README includes examples of retrieval traces for multi-hop questions

LLM usage20%

Prompts force evidence-backed synthesis and reject unsupported claims

Robustness10%

Handles overlapping policy sections, regional variants, and partial evidence

Correctness25%

Answers include all required facts and cite correct source IDs for 10/12 questions

Architecture20%

Retrieval supports multi-hop composition rather than only nearest-neighbour lookup

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

Build a multi-hop RAG system for policy questions