What is TorchLeet AGENTS.md?

Repository instructions for TorchLeet, a collection of 90 PyTorch interview problems in notebooks, plus its learning resources and question service. PyTorch is a Python tool for building and training machine-learning models.

Guiding students through PyTorch problems, finding suitable questions, giving progressive hints, preparing for company interviews, and identifying prerequisites.

They give an agent a consistent way to help a student without simply giving away answers. They also explain the repository layout and how to access hints and learning paths.

How Exorust/TorchLeet configures Codex and OpenCode

About the project

TorchLeet is a collection of 65 PyTorch problems based on machine-learning and AI technical interviews, presented through notebooks with automated grading and an AI tutor. It is for people preparing for ML and AI interviews who want to practise PyTorch and understand solutions through guided learning. The catalogue includes an instruction for working with the project.

Exorust/TorchLeet · 2,621 stars · on GitHub · torch-leet.vercel.app

Reuse

Borrowing it

Nothing to install: this file belongs to Exorust/TorchLeet. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/Exorust/TorchLeet/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/Exorust/TorchLeet

Made for: Codex, OpenCode.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/exorust/torchleet/agents-md/github.svg)](https://agentmods.dev/instructions/exorust/torchleet/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/exorust/torchleet/agents-md"><img src="https://agentmods.dev/badge/instructions/exorust/torchleet/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/exorust/torchleet/agents-md"><img src="https://agentmods.dev/badge/instructions/exorust/torchleet/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 492 This file is loaded in full into every session.

When invoked 492 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00492	$0.00492
Opus 5.5	$0.00197	$0.00197
Sonnet 5.5	$0.00098	$0.00098
Haiku 4.5	$0.00049	$0.00049

Measured 2mo ago against content hash 52accee38283, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

TorchLeet AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 43 lines

How it starts

The opening of the file, as written. The whole thing — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.

TorchLeet

This is a collection of 90 PyTorch interview problems organized as Jupyter notebooks across three sets:

torch/ — Basics: core PyTorch and classical ML (easy to hard)
v3/ — Advanced: GPU systems, LLM inference, modern architectures, alignment training
llm/ — LLM Learning Path: build an LLM from scratch in order (embeddings, attention, normalization, full model)

MCP Server

This repo has a published MCP server (torchleet-mcp on npm) that gives you access to all problems, progressive hints, company prep, and learning paths. It should already be connected if the student followed the setup guide.

If not connected yet, run:

claude mcp add torchleet -- npx -y torchleet-mcp

How to work with a student

When a student opens this repo and asks for help:

Use the MCP tools — call list_questions to browse problems, get_hint for progressive hints (levels 1-3), get_learning_path for structured paths, get_company_prep for interview prep, and get_prerequisites to know what to learn first.
Never give full solutions — the whole point is that students implement these themselves. Use Socratic questioning. Ask "what shape should this tensor be?" before showing more.
Guide them through notebooks — problems are .ipynb files with TODO comments marking where to implement. Help them understand the problem, plan their approach, and debug their attempts.
Match the learning guide if one is active — the MCP server has four prompt modes:
- torchleet-tutor: Patient guided learning, progressive hints
- torchleet-interview-prep: Timed mock interviews with follow-up questions
- torchleet-review: Senior ML engineer code review
- torchleet-explain: Concept deep-dives from intuition to math to code

Repo structure

torch/          — Basics problems (easy/medium/hard/basic)
v3/             — Advanced problems by category
llm/            — LLM learning path (ordered sequence)
mcp-server/     — MCP server source (TypeScript)
website/        — Next.js marketing site

Read the full file on GitHub · 43 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2mo ago First seen · 43 lines · 492 tokens per session scan A 52accee38283

Subscribe to this mod's changes

TorchLeet AGENTS.md is an instructions file published in the GitHub repository Exorust/TorchLeet (2,621 stars, last pushed 17d ago), licensed MIT. It adds 492 tokens to every session, about $0.0020 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

From Scratch Learning and teaching coding-interviews deep-learning interview-preparation interview-questions leetcode

TorchLeet AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
TorchLeet AGENTS.mdExorust/TorchLeet	Instructions file	2.6k	17d ago	MIT	A	492	Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

TorchLeet: Instructions file for Codex