Back to Tracks
TRACK 02

RUST AI
ENGINEER

Solve the world's most compute-intensive problems by engineering the next generation of AI infrastructure.

01

PYTHON INTEROP & PYO3

Move heavy lifting from Python to Rust. Creating native modules that outperform pure Python by 100x+.

PyO3MaturinZero-Copy
book

PyO3 User Guide

Official guide for writing Python extensions in Rust. Covers classes, functions, exception handling, and the GIL.

pyo3.rs
repo

Maturin — build and publish PyO3 modules

Tool for building and publishing Rust-based Python packages. Use alongside PyO3 to ship native extensions to PyPI.

github.com/PyO3/maturin
video

Rust + Python: Writing Native Extensions

Practical walkthrough of building a zero-copy Rust extension for Python using PyO3 and ndarray.

youtube.com — PyO3 tutorials
02

DEEP LEARNING IN RUST

Exploring the Candle and Burn ecosystems. Training and fine-tuning models without the Python overhead.

CandleBurnTensors
repo

Candle — Hugging Face ML framework for Rust

Minimalist ML framework by Hugging Face. Run LLaMA, Mistral, and other models entirely in Rust. Study all examples in the repo.

github.com/huggingface/candle
repo

Burn — deep learning framework

Flexible deep learning framework written in Rust. Supports training, inference, WASM, and multiple backends (NdArray, LibTorch, WGPU).

github.com/tracel-ai/burn
book

Burn Book — official documentation

Comprehensive guide to Burn: tensors, modules, training, and backend configuration.

burn.dev/book/
03

EDGE DEPLOYMENT

Deploying efficient models to WASM, IoT devices, and serverless runtimes. High-performance inference at the edge.

WASMONNXTensorRT
book

Rust and WebAssembly — official book

Build and deploy Rust to WASM targets. Covers wasm-bindgen, wasm-pack, and integration with JavaScript environments.

rustwasm.github.io/docs/book/
repo

tract — ONNX & NNEF inference in Rust

Pure Rust inference engine for ONNX models. No Python required — ideal for edge and embedded deployment.

github.com/sonos/tract
repo

rig — Rust AI agent framework

Build LLM agent loops and tool-calling pipelines in Rust. Read all examples, understand how tool execution and agent state are modelled.

github.com/0xPlaygrounds/rig
project

Build: LLM tool-execution service

Rust service that receives tool-call requests from an LLM via rig or raw OpenAI API, executes them in isolated Tokio tasks, returns results with timeouts.

rig.rs

Ready to compile?

Join 12,000+ engineers building the future of the web with Rust.