VLM & VFM Forward Deployed Engineer

About the role

Matroid is seeking a Forward Deployed Engineer specialized in Vision Language Models (VLM) and Visual Foundation Models (VFM) to design, deploy, and integrate cutting‑edge computer‑vision and multi‑modal AI solutions for industrial customers.

Key responsibilities

• Train and deploy state‑of‑the‑art vision‑centric and vision‑language models across manufacturing, automotive, logistics, aerospace, data‑center, security and other domains.
• Deploy end‑to‑end CV systems on cloud, edge and hybrid environments, defining benchmarks and iterating to meet production requirements for accuracy, latency, throughput and robustness.
• Design and build industrial‑grade imaging systems for consistent data collection and integrate Matroid’s platform into customer workflows such as MES, PLC, SCADA, QMS and video management systems.
• Act as the technical expert for scoping, model adaptation, deployment architecture, evaluation, integration and customer enablement, delivering workshops and deployment playbooks.

Required profile

• Bachelor’s degree in computer science, engineering, machine learning or a related technical field.
• Strong Python programming skills and experience building reliable production systems that interact with models, APIs, databases and customer infrastructure.
• Proven ability to rigorously evaluate AI systems, design benchmarks, analyze failure modes and improve performance.
• Excellent communication skills for both technical and commercial audiences.

Required skills

• Modern visual recognition models: object detection, segmentation, tracking, action recognition, anomaly detection, vision‑language models.
• Machine‑learning frameworks: PyTorch, TensorFlow, JAX.
• Tools and libraries: Hugging Face, NumPy, OpenCV.

What we offer

• Competitive salary and equity.
• Medical, dental and vision insurance with 100% premiums paid.
• Gym membership reimbursement, free lunch, healthy drinks and snacks.
• Flexible schedule, hardware/software budget and resources for continuous learning.