Senior or Staff Computer Vision Engineer

Full Time

|

San Francisco, CA

|

Hover

Hover helps people design, improve, and protect the properties they love. With proprietary AI built on over a decade of real property data, Hover answers age-old questions like “What will it look like?” and “What will it cost?” Homeowners, contractors, and insurance professionals rely on Hover to get fully measured, accurate, and interactive 3D models of any property—all from a smartphone scan in minutes.

At Hover, we’re driven by curiosity, purpose, and a shared commitment to serving our customers, communities, and each other. We believe the best ideas come from diverse perspectives and are proud to cultivate an inclusive, high-performance culture that inspires growth, accountability, and excellence.

Backed by leading investors like Google Ventures and Menlo Ventures, and trusted by industry leaders including Travelers, State Farm, and Nationwide, we’re redefining how people understand and interact with their spaces.

Why Hover Wants You

We’re expanding Hover’s 3D Reconstruction platform to solve one of our most ambitious technical challenges: transforming real-world sensor data into accurate, structured, CAD-quality models at scale.

Computer Vision doesn’t just support Hover’s products—it defines them.

We’re looking for either a Senior or Staff Computer Vision Engineer who can turn cutting-edge research into scalable production capabilities while helping shape the future of our reconstruction systems.

This role is ideal for someone who wants to own meaningful technical work end-to-end, move quickly without large-company bureaucracy, and help build accurate digital twin models of the real world at scale. You’ll work across research, experimentation, and production:

  • Reading papers

  • Reproducing and adapting promising academic approaches

  • Training deep learning models

  • Designing rigorous experiments

  • Building proof-of-concepts

  • Turning successful ideas into robust engineered systems that reach users

At the Staff level, this role is ideal for a technical leader who is equally comfortable with deep 3D reconstruction work, product strategy, architectural tradeoffs, experiment design, and production deployment.

You’ll help connect long-term technical vision, day-to-day engineering execution, and market-driven product needs so Hover’s computer vision roadmap delivers maximum customer and business impact.

You’ll join a high-performing team of computer vision researchers, reconstruction engineers, graphics specialists, and 3D modelers building systems that turn real-world capture into accurate digital twin models at commercial scale.

What You'll Do

In this role, you will design, prototype, evaluate, and productionize advanced computer vision and deep learning systems for:

  • 3D reconstruction

  • Scene understanding

  • Semantic modeling

  • Structured property representation

You will work across the full lifecycle of applied research and engineering, including:

  • Identifying relevant academic and industry approaches

  • Writing project plans and technical specifications

  • Building proof-of-concepts

  • Training and evaluating models

  • Analyzing reconstruction quality and performance

  • Integrating successful approaches into production systems

You will partner closely with:

  • 3D Reconstruction teams

  • Product

  • Design

  • Engineering

  • Modeling

  • Infrastructure

  • Graphics

  • Frontend and Backend teams

to turn ambiguous technical and customer needs into clearly scoped experiments and production capabilities.

Success in this role requires strong technical judgment, clear communication, and the ability to balance:

  • Research quality

  • User impact

  • Production constraints

  • Performance

  • Reliability

  • Cost

Your work may include:

  • Mobile capture

  • Aerial imagery

  • Multimodal sensor fusion

  • Multi-view reconstruction

  • Pose estimation

  • Feature matching

  • Dense geometry

  • Model fitting

  • Semantic understanding

  • Volumetric, surfel, or surface-based representations

  • CAD-quality structured outputs

  • Emerging methods such as Gaussian Splatting, VLMs, and foundation-model approaches for 3D understanding

Staff-Level Responsibilities

At the Staff level, you will also:

  • Shape technical direction across larger initiatives

  • Frame problems and size opportunities

  • Prioritize technical investments

  • Make "stop / iterate / scale" decisions across reconstruction approaches

  • Guide model development and data strategy

  • Define quality targets and product impact metrics

  • Communicate technical tradeoffs, resource needs, and roadmap implications to leadership

You will help raise the technical bar by:

  • Setting standards for performance, reliability, maintainability, and cost

  • Leading by example in Python and/or C++

  • Mentoring engineers through:

    • Design reviews

    • Architecture discussions

    • Experiment planning

    • Production-readiness reviews

Required Qualifications

  • 5+ years of experience in computer vision, machine learning, or 3D reconstruction in academic research and/or industry environments

  • Deep domain expertise and a strong track record of solving complex technical problems

  • Hands-on experience in one or more areas of modern 3D vision and reconstruction:

    • Multi-view geometry

    • Pose estimation

    • Camera calibration

    • Structure-from-Motion (SfM)

    • Feature matching

    • Dense reconstruction

    • Structured 3D reconstruction

    • Model fitting

    • Optimization

    • Scene understanding

    • Semantic modeling

    • Structured 3D representation

  • Practical deep learning experience applied to:

    • 3D reconstruction

    • Geometry

    • Correspondence

    • Pose

    • Segmentation

    • Semantic understanding

    • VLMs

    • Related spatial AI problems

  • Strong software engineering skills in Python and/or C++

  • Ability to prototype quickly, train models, evaluate approaches rigorously, and translate research into practical systems

  • Experience designing experiments with:

    • Clear metrics

    • Baselines

    • Datasets

    • Evaluation plans

    • Go / No-Go criteria

  • Ability to write:

    • Project plans

    • Technical specifications

    • Research summaries

    • Experiment reports

    • Production-readiness documentation

  • Proven technical ownership and ability to independently drive projects from ambiguity through delivery

  • Strong collaboration skills across technical and cross-functional teams

  • Strong product and engineering judgment

  • Master's or PhD in:

    • Computer Science

    • Machine Learning

    • Computer Vision

    • Robotics

    • Graphics

    • Related field

Additional Qualifications for Staff-Level Candidates

  • Experience building and shipping production CV, ML, or 3D reconstruction systems at scale

  • Ability to define:

    • Technical direction

    • Architecture strategy

    • Evaluation methodology

    • Quality and performance standards

  • Proven leadership of initiatives from research through production deployment and long-term ownership

  • Strong judgment balancing:

    • Technical ambition

    • Customer value

    • Scalability

    • Reliability

    • Cost

    • Business impact

  • Ability to operate effectively in ambiguous problem spaces

  • Strong communication and influence skills

  • Experience mentoring engineers and elevating engineering standards

  • Demonstrated ability to translate technical innovation into measurable product or business impact

Nice-to-Have Qualifications

  • Experience with modern reconstruction and spatial AI approaches such as:

    • Gaussian Splatting

    • Neural reconstruction

    • DUSt3R

    • MASt3R

    • VGGT

    • RoMa

    • Surfels

    • Volumetric methods

    • VLMs

    • Foundation-model approaches for 3D understanding

  • Publications in venues such as:

    • CVPR

    • ICCV

    • ECCV

    • 3DV

    • ISMAR

    • SIGGRAPH

    • NeurIPS

    • ICLR

  • Experience with large-scale production ML systems:

    • Model serving

    • Monitoring

    • MLOps

    • Distributed training

    • Data pipelines

    • Cost and performance optimization

  • Familiarity with:

    • GCP

    • Kubernetes

    • Ray

    • Large-scale data processing frameworks

  • Experience with:

    • Geometry processing

    • Graphics

    • Rendering

    • CAD/BIM workflows

    • Mesh processing

    • Interactive 3D systems

  • Experience with PyTorch or TensorFlow

  • Practical experience using AI-assisted development tools such as:

    • Cursor

    • Claude Code

    • GitHub Copilot

    • Similar productivity tools

Benefits

Compensation

  • Competitive salary

  • Meaningful equity in a fast-growing company

Healthcare

  • Comprehensive medical, dental, and vision coverage for employees and dependents

Paid Time Off

  • Unlimited and flexible vacation policy

Paid Family Leave

  • Generous paid parental leave and child bonding leave

Self-Care Days

  • One dedicated self-care day each month

Remote Wellbeing Resources

  • Fitness classes

  • Meditation and mindfulness tools

  • Virtual therapy

  • Family planning assistance

Learning & Development

  • Support for management training

  • Conferences

  • Workshops

  • Certifications

Hybrid Work Model

Hover has hubs in:

  • New York City

  • San Francisco

Employees living within a 50-mile radius of either office are expected to work from the office at least three times per week to foster collaboration and connection.

At this time, Hover is not considering fully remote roles.

Compensation Range

The U.S. base salary range for this full-time position is:

$188,000 – $322,000 annually

Salary is determined based on:

  • Role and level

  • Geographic location

  • Relevant experience

  • Skills

  • Education and training

The posted range spans multiple levels to reflect the breadth of candidates being considered.