11 months ago
Base Salary
$150k - $220k/yr
Responsibilities
- Improve the accuracy and latency of core action models across diverse web applications.
- Design and optimize LLM inference pipelines, including token caching, streaming architectures, and network-level client-server optimizations.
- Build evaluation frameworks and data pipelines to measure and improve model quality at scale.
- Develop synthetic data generation pipelines for browser-interaction training data.
- Use DOM states, accessibility trees, and user-interaction data to improve browser understanding.
- Experiment with retrieval-augmented approaches using vector databases for contextual memory.
- Ship end-to-end ML features directly to users.
Requirements
- Hands-on experience training and deploying machine learning models in production.
- Experience optimizing inference pipelines for very low latency.
- Experience with LLMs, transformer architectures, or sequence prediction problems.
- Ability to work across a system involving a Chrome extension, Electron app, Cloudflare Workers edge proxy, and inference providers.
- Strong attention to data quality and experience building tooling to measure and improve it.
- Experience with browser automation, Chrome extensions, or web scraping at scale is a bonus.
- Familiarity with accessibility trees and DOM parsing for page understanding is a bonus.
- Background in reinforcement learning or online learning from user interaction data is a bonus.
- Experience with vector databases such as Turbopuffer or Pinecone and hybrid search is a bonus.
- Full-stack development experience with TypeScript, Node.js, and React is a bonus.
Tech Stack
Categories
About Composite
Composite builds a proactive browser automation platform that predicts and executes next steps in web-based workflows for professionals. Its product runs in the browser, clicking, typing, and navigating based on page context and prior interactions, and is offered as a SaaS productivity tool. Privately held and founded in 2024, the company is headquartered in San Francisco.