About the Role
Join an early-stage industrial AI team building wearable AI assistants for field technicians. You will own key parts of the multimodal AI stack, taking vision-language systems from development into real-world workflows.
What You'll Do
Build and ship agentic vision-language pipelines that use visual reasoning and tools to support industrial workflows.
Develop evaluation systems, capture failure modes, and create data feedback loops to improve model quality.
Adapt inference and model orchestration for changing connectivity, latency requirements, and edge deployment.
Create real-time voice and video interfaces, including conversational and proactive assistance.
Fine-tune open-weight models using techniques such as supervised fine-tuning, preference optimization, and quantization.
What We're Looking For
Zero to three years of experience shipping production vision-language or multimodal AI systems, with end-to-end ownership of models and orchestration.
Strong Python and PyTorch skills, plus hands-on experience with tools such as vLLM, Triton, Ray Serve, ONNX, TensorRT, Hugging Face Transformers, and Docker.
Experience building multi-step agent loops with genuine tool use, evaluation harnesses, and data improvement workflows.
Applied experience with model fine-tuning and evaluation, rather than a focus solely on pretraining from scratch.
Background in computer vision, wearables, industrial AI, autonomous systems, or related applied domains; edge inference or on-prem deployment experience is valuable.
A relevant degree or equivalent evidence of strong technical fundamentals and shipped AI systems.
Compensation & Benefits
Base salary is $180,000 to $230,000 annually. Visa sponsorship is available.
Location
On-site in San Francisco, United States.