Robotics Research Engineer, Video Perception & Auto-labeling

Miraxis AI builds AI-assisted data generation systems for Physical AI and robotics. They are seeking a Robotics Research Engineer for Video Perception & Auto-labeling to enhance their video annotation pipeline by developing visual evidence layers and ensuring the reliability of perception artifacts.

Responsibilities:

• Integrate promptable segmentation and tracking models into the annotation evidence pipeline
• Produce masks, boxes, tracks, object IDs, confidence scores, and evidence references for hands, tools, objects, surfaces, fixtures, and state-changing entities
• Integrate dense visual feature models for visual similarity, mask-quality support, object and scene features, out-of-distribution checks, and transition-evidence support
• Integrate video-text embedding models for clip search, inverse video search, semantic deduplication, scenario mining, hard-negative mining, and gold-sample candidate selection
• Build durable evidence artifacts that attach perception outputs to claim-level annotations
• Create uncertainty and failure flags for occlusion, hidden hands, weak track continuity, fast motion, camera shake, object confusion, poor visibility, track drift, and ambiguous contact
• Build evidence packaging for reviewers, model evaluation, annotation correction, and downstream audit workflows
• Support reviewer workflows so humans can inspect, correct, or reject masks, tracks, and evidence spans efficiently
• Build visualization and review hooks using Rerun, Foxglove or equivalent tooling
• Work with world-model engineers to provide masks, tracks, features, and embeddings for transition-evidence and latent-residual experiments
• Work with platform engineers to version every model output, including checkpoint, config, preprocessing, prompt or query, frame range, artifact hash, and storage reference
• Ensure perception and embedding outputs remain evidence and routing features only. They should not automatically determine annotation truth or replace human review for high-risk cases