3D Reconstruction Scientist

Array Labs
Redwood City, United States
3 days ago
Apply on startup.jobs
Prepare application

Role details

Contract type
Permanent contract
Employment type
Full-time (> 32 hours)
Experience level
Experienced
Experience required
2 years minimum
Working hours
Regular working hours
Job source

Tech stack

Geographic Information Systems Artificial Intelligence Computer Vision C++ (Programming Language) Computer Clusters Geospatial Intelligence Python (Programming Language) MATLAB Machine Learning Signal Processing Information Technology

Job description

Array is building that system. We are developing a coordinated fleet of radar satellites designed to create high-resolution 3D data products of the Earth at unprecedented scale, speed, and reliability. Our technology is built to work across wide areas, through clouds, day and night, and to support both commercial and government customers operating in environments where timely physical-world understanding matters.

We design and build our satellites, radar payloads, sensing systems, and data products end-to-end. Our goal is to make 3D geospatial intelligence dramatically more accessible, useful, and operationally relevant for customers in mapping, infrastructure, disaster response, defense, intelligence, energy, telecommunications, mining, and other industries that depend on understanding the changing Earth. About the Job As a 3D Reconstruction Scientist on this team, you will develop and deploy algorithms that reconstruct 3D structure from distributed radar and optical sensors. You’ll work on multi-view geometry, tomography, and large-scale inverse problems, and you’ll be responsible for taking methods from prototype to production on real satellite data. The approaches you design will directly set the limits on the resolution, accuracy, and reliability of Array’s 3D products. As part of our quickly growing engineering team, you will play a critical role in the design of the world’s most advanced radar satellite constellation-one that will significantly improve humanity’s ability to rapidly and comprehensively understand our changing planet., * Research, design, and implement tomographic algorithms for 3D image reconstruction utilizing complex phase-history data from radar and optical sensors

  • Develop autofocus techniques to sharpen images affected by inaccurate position knowledge of the sensor platform, and/or phase inaccuracies caused by various hardware imperfections in our distributed sensor system
  • Develop efficient data preprocessing and calibration techniques to enhance the quality and accuracy of 3D reconstructions
  • Optimize algorithms for near real-time production of 3D imagery in our cloud-based production pipeline
  • Collaborate with cross-functional teams to integrate 3D image reconstruction solutions into our products and services
  • Conduct thorough testing, validation, and performance evaluation of 3D image reconstruction algorithms to ensure they meet end-user expectations

Requirements

  • Bachelor’s in Computer Science, Electrical Engineering, or a related field
  • 4+ years of experience in 3D image reconstruction, 3D computer vision, multi-view geometry or a related field
  • Excellent teamwork and communication skills
  • Learns new concepts rapidly, completely, and in a self-directed manner
  • High levels of self-motivation and personal accountability
  • Ability to work in a fast-paced environment under significant time constraints
  • Proficient in Python, MATLAB, or C/C++, * Master’s degree in Computer Science, Electrical Engineering, or a related field
  • 2+ years of experience deploying 3D image processing/generation algorithms into production pipelines
  • 2+ years of experience in one or more of the following areas:
  • signal processing for tomographic SAR and/or interferometric SAR
  • Depth estimation using stereo images, multi-view geometry, or optical flow
  • Generating 3D images from 2D sensors.
  • Estimation theoretic approaches to 3D localization
  • Ambiguity resolution for 3D localization
  • Experience adapting image processing algorithms for GPU clusters
  • Experience with stochastic and adaptive signal processing techniques
  • Experience extracting insights from 3D data products via AI/ML techniques
  • Experience creating 3D data products from geospatial data sets is a plus

Benefits & conditions

  • Array Labs is an Equal Opportunity Employer. Employment decisions are made on the basis of merit, competence, and job qualifications and will not be influenced in any manner by gender, color, race, ethnicity, national origin, sexual orientation, religion, age, gender identity, veteran status, disability status, marital status, mental or physical disability or any other legally protected status. Interview Process We will conduct interviews via Google Meet; the typical process takes around 2-4 weeks to complete from start to finish. Hiring and Compensation Strategy Our hiring and compensation strategy is simple: 1) find uncommonly good people 2) pay them uncommonly well You can anticipate competitive pay, with high flexibility between salary and equity-based compensation. Why Join Array Labs? Array Labs is launching a constellation of satellites to create the first high-resolution, real-time, three-dimensional model of Earth. Our next-generation satellite technology will offer image quality 60x greater than traditional techniques, profoundly expanding humanity’s ability to understand and respond to events on a global scale. In forging an affordable, accessible, accurate representation of Earth, our work has the potential to transform the face of dozens of fields, including autonomy, telecommunications, disaster relief, gaming, climate science, defense and construction.

Apply for this position

This job is hosted externally. Click below to view the full posting and apply.

Apply on startup.jobs
Prepare application

Good distractions

Talks and stories from around this role — technically off-topic, practically not.

3:14 min

Structuring career paths and localized data architectures

Ulrich Wurstbauer +1 · LIVE

2:36 min

Applying supervised machine learning for practical rule extraction

Katja Träumner

5:55 min

Practical applications and use cases for computer vision

Flo Pachinger · LIVE

2:40 min

Motivations for transitioning legacy MATLAB repositories to Python

Michael Niebisch Michael Niebisch · World Congress 2024

2:56 min

Introduction of panelists and their roles in AI engineering

Damandeep Kochhar Damandeep Kochhar +4 · World Congress 2026 Europe

1:57 min

Evolution of machine learning algorithms and computing hardware

Alexandra Waldherr · LIVE

Videos

See all

Related articles

See all