Sr. AI Researcher, Agentic Systems
About The Position
The Sr. AI Research Engineer will build agentic systems that reason across OpenSpace’s site data, including 360° imagery, point clouds, BIM models, drawing sets, specifications, and newer data streams from mobile capture, wearables, and field audio. The role builds on computer vision perception primitives, combining them with frontier models to answer practical questions from project teams and produce verifiable answers that users can act on. This is applied research focused on deciding what an agent should see in a four-dimensional site record, designing the tools it can call, and evaluating whether its results are reliable enough to ship. The engineer will develop retrieval and grounding over spatially and temporally indexed site data, create evaluations for tasks with ambiguous ground truth, select models based on cost, latency, and accuracy, and make outputs traceable to supporting evidence. The role also involves rapid prototyping and partnering with platform engineers to bring successful systems into production.
Requirements
- M.S., Ph.D. in a related field, or equivalent industry experience.
- At least 5 years of experience building ML systems.
- Experience taking at least one LLM- or VLM-based system into production and maintaining it.
- Experience designing evaluations from scratch when ground truth is expensive, sparse, or contested.
- Strong multimodal fluency, including vision-language models, embeddings and retrieval, context design, and fine-tuning when appropriate.
- Strong engineering skills, including Python, asynchronous orchestration, tracing, and cost instrumentation.
- Comfort with spatial data; coordinate frames, camera poses, and projections should be familiar.
Nice To Haves
- Post-training or reinforcement learning on agent trajectories.
- Document understanding for technical drawings, specifications, or other structured PDFs.
- Experience with sensor streams beyond core imagery, such as mobile capture, wearables, or field audio.
- Experience in construction, robotics, geospatial, or another domain where models interact with the physical world.
Responsibilities
- Design and build agent systems that plan and execute multi-step tasks across imagery, geometry, and documents.
- Build retrieval and grounding over a spatially and temporally indexed site record so agents receive relevant evidence.
- Design the tools and primitives agents use to locate components, compare captures, read drawing regions, and query BIM elements.
- Build evaluations for tasks with ambiguous ground truth and assess what is ready for production.
- Select appropriate models for each task, considering cost, latency, and accuracy at production volume.
- Make outputs traceable to the supporting pixel, drawing sheet, or specification section.
- Prototype rapidly, discontinue approaches that do not work, and partner with platform engineers to ship successful systems.
Benefits
- 100% employer-paid medical, dental, and vision coverage.
- Flexible paid time off.
- 401(k) with company match.
- Paid parental leave.
- Home office stipend.
- Employee referral program.
- Annual company retreats and team gatherings.
- Equity awards.
- Region-specific health and wellness benefits.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Senior
