🛠 Tech Stack

💼 About This Role

You'll design the ML strategy for extracting intelligence from LiDAR and imagery data at a Boston-based govtech startup. Your work will turn millions of images and point clouds into structured geospatial knowledge, directly impacting public infrastructure for cities nationwide. You'll be the first ML hire with full ownership of the learning system from scratch.

🎯 What You'll Do

• Design and own the ML strategy for extracting intelligence from imagery and LiDAR data
• Build pipelines to detect, classify, and measure infrastructure assets across millions of images
• Develop cross-modal representations encoding visual and geometric information
• Ship models to production alongside infrastructure engineers

📋 Requirements

• 4-8 years of professional ML experience with at least 2 years on large-scale sensor data
• Production experience with data-efficient learning (active learning, auto-labeling, or semi-supervised)
• Hands-on experience with LiDAR-camera fusion or multi-modal perception
• Experience training and evaluating models on millions of images

✨ Nice to Have

• Experience at an autonomous vehicle company (perception, mapping, or data engine)
• Familiarity with point cloud architectures (PointNet, voxel-based, or multi-view)
• Experience with vision foundation models (SAM, CLIP, DINOv2) and adapters

🎁 Benefits & Perks

• 💰 Competitive salary and equity ($180K-$350K total comp, equity-heavy)
• 🏖️ Generous PTO and paid parental leave
• 🏥 Comprehensive health, dental, and vision coverage
• 📚 Annual professional development stipend
• 🏦 401(k) with immediate vesting