Overview

Weco: The Self-Improving Code Platform

Weco systematically optimizes your code, guided directly by your evaluation metrics.

The Weco CLI, powered by our core engine AIDE, leverages a tree search approach guided by LLMs to iteratively explore and refine your code. It automatically applies changes, runs your evaluation script, parses the results, and proposes further improvements based on the specified goal.

An example of the tree search process is shown below:

Key Applications

Weco can be applied to a wide range of code optimization tasks, including:

• GPU Kernel Optimization: Reimplement PyTorch functions using CUDA or Triton optimizing for latency, throughput, or memory_bandwidth.
• Model Development: Tune feature transformations or architectures, optimizing for validation_accuracy, AUC, or Sharpe Ratio.
• Prompt Engineering: Refine prompts for LLMs, optimizing for win_rate, relevance, or format_adherence.

What's Next?

• Installation
• Getting Started
• CLI Reference
• Writing Good Evaluation Scripts
• FAQ
• Examples