Founding Engineer, Research

Cua is building infrastructure for safely and scalably running general AI agents on real computers and apps. With 9k+ GitHub stars in just 4 months and backing from Y Combinator, we're advancing agentic AI from research to production. We're hiring a Research Engineer to help develop and scale advanced multi-modal AI agents — working across model training, agent benchmarking, and real-world deployment. ⸻ WHAT YOU'LL DO You'll sit at the intersection of applied research and engineering, helping us build, evaluate, and ship the next generation of generative agent systems. Example work includes: • Designing and running experiments on commercial/open-source LLMs (e.g., OpenAI, LLaMA, Qwen) • Building scalable pipelines for training, fine-tuning, and evaluating multi-modal agents • Developing tools and benchmarks to test agent reasoning, control, and performance • Improving infrastructure for deploying agentic AI models across OS environments • Supporting research-to-production workflows for internal and external users ⸻ WHAT WE'RE LOOKING FOR • Strong experience with generative AI, LLMs, and agentic systems • Hands-on with commercial and open-source models at scale • Proficiency in Python and PyTorch (C++/Java a plus) • Experience with data curation pipelines and multi-modal training workflows • Comfortable designing experiments, testing infra, and pushing code into prod • Exposure to cloud compute (AWS/GCP), APIs, structured/unstructured data • Open-source or competition experience is a plus ⸻ LOGISTICS • Full-time, remote-friendly (SF-based team preferred) • Role blends fast-paced engineering with cutting-edge research • Work used by thousands of developers building with Cua ⸻ APPLY • CV + GitHub or portfolio • Short note on a project or experiment you've recently led 🌐 trycua.com We're committed to building a diverse, inclusive team — all backgrounds welcome.

Cua is building the infrastructure that lets general AI agents safely and scalably use Computers and Apps like humans do. With 9k+ GitHub stars in just 4 months and a seed round closed, we're providing: An open-source framework for building and evaluating general-purpose AI agents. A cloud container platform for sandboxed, scalable agent execution environments. A blueprint for what production-grade general agent systems should look like - backed by research.