Job description

THE ROLE At Mach9, ML Engineers build the perception models at the core of our AI-enabled CAD system. We build models to extract 3D object and line features from dense LiDAR point clouds and imagery. Our unique data advantage allows us to develop and train cutting edge 3D scene understanding models that serve real surveyors and engineers in the field. This role is product-focused. You will be working closely with customers to develop CV/ML pipelines and workflows that solves their problems. Your responsibility is to ship new features and product lines end-to-end from data strategy, to model training, to evaluation. This role is ideal for early-career engineers (new grad to ~3 years) who learn fast, are curious about how things work, and get real satisfaction from shipping. You need to be able to come up with ideas, try them, get feedback and iterate quickly. RESPONSIBILITIES - Ship new extraction features end to end: scoping with product and customer-facing teams, data and labeling strategy, model selection or fine-tuning, evaluation, and integration into Digital Surveyor. - Build on existing model families and open-source tooling first (vision foundation models, VLMs, point-cloud backbones, classical geometry), and know what's out there and what each is good for. - Adapt our production models to new object classes, regions, and sensor types as customers bring them. - Build evaluation pipelines that measure improvements and regression alike. - Iterate on feature extraction pipelines according to customer feedback and metrics changes. REQUIREMENTS - BS or MS in Computer Science, EE, Robotics, or a related field, or equivalent experience. - Strong coding abilities (preferably Python + Pytorch), and comfortable working with a large codebase. - Hands-on experience training or fine-tuning a vision model (segmentation, detection, or 3D) through work, internships, or projects. - Working knowledge of geometry for 3D perception: coordinate systems, transforms, projecting between images and point clouds. - Curiosity and speed: you pick up new frameworks and papers quickly and can explain what you learned to teammates. - Clear communicator with engineers, product, and the surveyors who use what you build. - Fluent with AI coding assistants. BONUS QUALIFICATIONS - Have used open-source vision foundation models (SAM family, DINO, Grounding DINO, or similar) in a task before. - Experience with point clouds or LiDAR data (Open3D, PDAL, PyTorch3D, sparse convolution libraries). - Experience deploying models to production: ONNX or TensorRT, batch inference on cloud GPUs.

Requirements

Must-haves

Ship new extraction features end to end: scoping with product and customer-facing teams, data and labeling strategy, model selection or fine-tuning, evaluation, and integration into Digital Surveyor.
Build on existing model families and open-source tooling first (vision foundation models, VLMs, point-cloud backbones, classical geometry), and know what's out there and what each is good for.
Adapt our production models to new object classes, regions, and sensor types as customers bring them.
Build evaluation pipelines that measure improvements and regression alike.
Iterate on feature extraction pipelines according to customer feedback and metrics changes.
BS or MS in Computer Science, EE, Robotics, or a related field, or equivalent experience.
Strong coding abilities (preferably Python + Pytorch), and comfortable working with a large codebase.
Hands-on experience training or fine-tuning a vision model (segmentation, detection, or 3D) through work, internships, or projects.
Working knowledge of geometry for 3D perception: coordinate systems, transforms, projecting between images and point clouds.
Curiosity and speed: you pick up new frameworks and papers quickly and can explain what you learned to teammates.
Clear communicator with engineers, product, and the surveyors who use what you build.
Fluent with AI coding assistants.
Have used open-source vision foundation models (SAM family, DINO, Grounding DINO, or similar) in a task before.
Experience with point clouds or LiDAR data (Open3D, PDAL, PyTorch3D, sparse convolution libraries).
Experience deploying models to production: ONNX or TensorRT, batch inference on cloud GPUs.

Nice-to-haves

None flagged.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

mlaipython

Platform: ashby
Effort: quick

Benefits

Benefits detected in the listing.

Vision