Vision AI & 3D ML Research Engineer (Remote)

Centific; Seattle, Northern (WA, KY); Hybrid; USD 180,000 - 240,000; Full time

Job summary

Centific seeks an AI Research Engineer to push the frontier of Vision AI, VLMs, and embodied/physical AI, translating cutting‑edge research into production systems that perceive, reason, and act in the real world.

You will own high‑leverage experiments from paper → prototype → deployable module in our platform, collaborating with the Vision AI and Physical AI Robotics teams to build 3D reconstruction and perception capabilities using state‑of‑the‑art tools such as Omniverse and Isaac Sim.

Qualifications

• Masters/Ph.D in CS/EE/Robotics with active publications in CV/ML/Robotics.
• Strong PyTorch and Python; familiarity with CUDA profiling and mixed precision training.
• Proven ability to move from paper to code to reproducible results with rigorous experimentation.
• Experience with 3D perception, VLMs, and embodied/physical AI is a plus.

Responsibilities

• Advance Visual Perception: build and fine‑tune 2D/3D detection, tracking, segmentation, and scene understanding.
• Multimodal Reasoning: train/evaluate VLMs and design retrieval‑augmented and agentic loops.
• Physical AI & Embodiment: prototype perception‑in‑loop policies linking pixels to actions.
• Data & Evaluation at Scale: curate datasets, define KPIs, and run ablations.
• Systems & Deployment: package research into reliable services on Kubernetes, Docker, Ray, FastAPI.
• Agentic Workflows: orchestrate multi‑agent pipelines integrating perception, reasoning, and code generation.

Skills

PyTorch

Python

CV/ML research

Publications in CV/ML/Robotics

Education

Masters/Ph.D in CS/EE/Robotics

Tools

ROS2

TensorRT

ONNX

Omniverse

Job description

Centific seeks an AI Research Engineer to push the frontier of Vision AI, VLMs, and embodied/physical AI, translating cutting‑edge research into production systems that perceive, reason, and act in the real world.

You will own high‑leverage experiments from paper → prototype → deployable module in our platform, collaborating with the Vision AI and Physical AI Robotics teams to build 3D reconstruction and perception capabilities using state‑of‑the‑art tools such as Omniverse and Isaac Sim.