AI Research Scientist

Sourced from centific.wd1.myworkdayjobs.com

Centific is hiring an AI Research Engineer to advance Vision AI, vision-language models, and physical/embodied AI. You will own high-leverage experiments from paper to prototype to deployable module across 2D/3D perception, multimodal reasoning, and robotics simulation. The role blends top-tier research rigor with pragmatic product focus, working alongside a Principal Architect and senior researchers.

• Build and fine-tune models for detection, tracking, 2D/3D segmentation, pose/activity recognition, and scene understanding
• Train and evaluate vision-language models for grounding, dense captioning, temporal QA, and agentic perception-action loops
• Prototype perception-in-the-loop policies bridging simulation and real data for manipulation, navigation, and safety workflows
• Curate datasets, author evaluation protocols/KPIs, and run rigorous ablations
• Package research into reliable services with profiling, telemetry, and CI for reproducible science
• Orchestrate multi-agent pipelines combining perception, reasoning, simulation, and code generation

• Masters/PhD in CS/EE/Robotics or related field with active publications (CVPR/ICCV/ECCV, NeurIPS/ICML/ICLR, CoRL/RSS)
• Strong PyTorch or JAX and Python skills, including CUDA profiling and mixed-precision training
• Demonstrated research in computer vision and at least one of: VLMs, embodied/physical AI, or 3D perception
• Proven ability to move from paper to code to ablation to result with rigorous experiment tracking

• Experience with video models (TimeSFormer/MViT/VideoMAE), diffusion, or 3D GS/NeRF pipelines
• SLAM or scene reconstruction background
• Multimodal grounding or temporal reasoning work
• Familiarity with ROS2, DeepStream/TAO, or edge inference (TensorRT, ONNX)
• Scalable training with Ray, distributed data loaders, sharded checkpoints
• Public code artifacts and first-author publications