# Exorust/TorchLeet

LeetCode for PyTorch — 65 ML/AI interview problems from real interviews at Google, Meta, Anthropic. Jupyter notebooks, an auto-grader, and an MCP AI tutor.

Repository: https://github.com/Exorust/TorchLeet
Canonical: https://ross.abutalabs.com/products/torchleet
Homepage: https://torch-leet.vercel.app
Language: Jupyter Notebook
License: MIT
License Family: permissive
Topics: coding-interviews, deep-learning, from-scratch, interview-preparation, interview-questions, leetcode, llm, machine-learning-interview, pytorch, rlhf, transformers, triton
Last push: 2026-08-19T20:41:12+00:00

## Health v2 (maintenance only)
Score: 65/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 35, longevity 44
- inputs: {"age_days": 617, "days_push": 14, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2430, forks 305 (observed 2026-08-28T04:06:50.961937+00:00)

## What it is
TorchLeet is a collection of 65 PyTorch practice problems drawn from real ML/AI interviews at companies like Google, Meta, and Anthropic, delivered as Jupyter notebooks with an auto-grader. It includes an MCP server that turns AI assistants into a no-spoilers PyTorch interview tutor with hints, mock interviews, and learning paths.

## Use cases
- prepare for ML/AI coding interviews with PyTorch problems
- practice implementing transformers, attention, and RLHF from scratch
- build an LLM from scratch as a guided learning path
- get AI-tutored PyTorch practice with progressive hints via MCP
- practice problems asked at Google, Meta, Anthropic, and OpenAI interviews
- check my PyTorch solutions automatically with an auto-grader

## When to choose
- you are preparing for machine learning or AI engineer interviews and want hands-on PyTorch practice
- you want to deeply understand PyTorch internals by implementing things from scratch without LLM help
- you want a structured curriculum covering basics through advanced topics like Triton kernels, KV cache, and DPO
- you want an AI tutor that guides without spoiling solutions

## When to avoid
- you need a production ML library or framework rather than practice exercises
- you are a complete beginner to Python or deep learning with no PyTorch basics
- you want quick answers generated by an LLM rather than struggling through problems yourself

## Facets
- artifact type: learning-resource
- maturity: active
- function: machine-learning, deep-learning, llm-training, prompt-engineering, mcp, testing
- domain: machine-learning, deep-learning, large-language-models, education, tutorials
- platform: python, cross-platform
- tags: pytorch, interview-preparation, leetcode, jupyter-notebooks, auto-grader, rlhf, transformers, triton, coding-interviews, mcp-server, web

## Member repositories
- Exorust/TorchLeet (main) score 65

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:50.961937+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:31:25.738702+00:00, confidence not recorded.
  - readme: https://github.com/Exorust/TorchLeet (fetched 2026-08-28T04:06:50.961937+00:00, sha 4936761a5a92)
  - homepage: https://torch-leet.vercel.app (fetched 2026-08-29T10:12:42.349415+00:00, sha 4034b4ba8513)
  - registry_pypi: https://pypi.org/pypi/torchleet/json (fetched 2026-08-29T10:12:42.352139+00:00, sha c2744b0acd4a)
- Data as of 2026-08-30T08:39:29.467469+00:00.