Founding Engineer — Open-Source AI Deployment Tools

This role is for someone who understands how open models are served, fine-tuned, evaluated, and deployed in real-world environments — but you do not need to be an expert in every area. We care more about strong engineering ability, curiosity, and the willingness to learn quickly.

Responsibilities

• Build developer tools that make it easier to deploy, serve, monitor, and improve open-source AI models.
• Work with modern model-serving technologies such as vLLM, TGI, Ollama, llama.cpp, TensorRT-LLM, or similar systems.
• Help design workflows for fine-tuning, evaluation, benchmarking, and model comparison.
• Integrate with popular open-source AI ecosystems, including tools for inference, vector search, observability, datasets, and model hosting.
• Experiment with open models from ecosystems such as Hugging Face, Meta Llama, Mistral, Qwen, DeepSeek, and others.
• Build reliable backend systems, APIs, and infrastructure that make complex AI workflows feel simple for users.
• Work closely with users to understand pain points around deploying and improving open models.
• Move quickly from idea to prototype to product, especially in areas that are still evolving.
• Contribute to early technical decisions, product direction, and engineering culture as part of the founding team.