About this role
The ideal candidate for this role enjoys tackling complex, real-world problems where technology directly impacts public good. This position is within a fast-moving team in Boston that is developing physical AI solutions to manage and improve public infrastructure. Cyvl is focused on creating intelligence from physical data, using sensors and computer vision to monitor the real world. This work directly contributes to safer commutes and better public spending by providing actionable data on infrastructure condition.
As an AI Engineer, the focus will be on transforming vast proprietary datasets—including LiDAR and imagery—into functional intelligence. This involves developing computer vision models to assess physical assets and creating sophisticated LLM-powered agents that allow customers and partners to query the physical world directly. The role spans the entire lifecycle, from initial prototype development through to deploying production-grade models that customers use daily.
This role is designed for someone who thrives on ownership, is comfortable operating without a fully defined playbook, and is driven by the outcome of building tangible solutions. Individuals who enjoy bridging the gap between complex data science and practical, deployed software systems will find this environment highly suitable.
What you will do
- Develop and enhance computer vision and 3D perception models to detect, classify, and assess infrastructure elements such as pavement, signs, sidewalks, and trees from sensor data.
- Construct LLM-powered products, including agentic workflows, the MCP server, and AI copilots designed to integrate Cyvl data into customer tools.
- Assume full ownership of models, managing them from data and training stages through to evaluation, deployment, and ongoing production monitoring.
- Design rigorous evaluation sets and metrics that accurately reflect the actual needs of the end-users.
- Continuously explore and integrate new state-of-the-art tools into both the product development and internal workflows.
What you need
- Possesses at least three years of experience in building and shipping Machine Learning or AI systems in a production environment.
- Demonstrates strong proficiency in Python and a modern deep learning framework, with a preference for PyTorch.
- Requires hands-on experience with computer vision, 3D perception, or applications involving LLMs, such as RAG, tool use, agents, or evaluations.
- Must have solid software engineering fundamentals, including experience with testing, code review processes, and cloud deployment.
- The ability to rapidly move from an ambiguous problem definition to a working prototype and subsequently develop it into a production-grade system is essential.
Nice to have
- Point cloud or multimodal work (LiDAR and imagery fusion)
- Model Context Protocol (MCP), foundation model APIs, or agent frameworks
- Geospatial or remote sensing background
- MLOps: model serving, GPU infrastructure, experiment tracking
- Advanced degree in CS, ML, robotics, or a related field, or equivalent experience
Benefits
- Equity
- Health, dental, and vision coverage for the employee and family
- 401k with a real match, vested day one
- Generous PTO & Paid Parental leave
- Learning stipend
- Paid to live close (monthly bonus for living near the Boston office)
- Team events in Boston
Skills and tech stack
Questions candidates ask
What is the remote work policy?
This role is based in Somerville, MA, or remote from San Francisco, CA. If a candidate is based in San Francisco, they will be with the team in Somerville for moments that matter like onboarding, kick-offs, and milestones.
What is the compensation structure?
The salary range for this position is listed as $125K – $225K, and the compensation also includes equity.
What is the primary tech stack used?
The required skills include Python and a modern deep learning framework, with PyTorch being preferred.
What is the team culture like?
The team operates with a culture that emphasizes intense work coupled with care for one another, genuine fun, and a shared belief in the work being done.