Simplify Logo
CoVar

CoVar

Custom AI/ML solutions for mission-critical sectors

Machine Learning Engineer

Full-Time
No salary listed
Senior
Bachelor's, Master's, PhD
McLean, VA, USA
Hybrid

Hybrid role with occasional travel to Durham, North Carolina, and customer sites.

No H1B Sponsorship
US Citizenship, US Top Secret Clearance Required

About the job

Requirements
  • At least 5 years of experience designing, building, and deploying machine learning systems in production.
  • A strong track record leading technical delivery for complex projects.
  • Deep understanding of machine learning fundamentals, including gradient descent, cross-validation, ROC/PR curves, confusion matrices, and mean average precision.
  • Computer vision experience with modern architectures such as the YOLO family, CenterNet, RetinaNet, Detectron2, vision transformers, and segmentation networks, including augmentation strategies and evaluation.
  • Experience with data curation and annotation workflows and dataset quality control.
  • Proficiency in Python, NumPy, SciPy, pandas or Polars, PyTorch or TensorFlow, Git, continuous integration and continuous delivery pipelines, automated testing, and code quality practices.
  • Experience with experiment tracking, Docker, ONNX or TensorRT, and deploying inference services to edge devices such as NVIDIA Jetson.
  • Excellent technical communication, customer-facing experience, and the ability to lead cross-functional efforts.
  • A B.S. degree in engineering, mathematics, computer science, or a related field.
  • Eligibility for a U.S. security clearance and U.S. citizenship.
Responsibilities
  • Scope requirements, set technical strategy and milestones, plan resourcing, manage risks, and deliver results for Department of Defense machine learning and computer vision projects.
  • Run technical discussions, requirement discovery, demonstrations, standing meetings, and briefings with senior stakeholders; translate feedback into clear roadmaps.
  • Build data pipelines, train and evaluate computer vision models, and write production code.
  • Design, train, and optimize detection, segmentation, and tracking models and custom computer vision pipelines while addressing imbalanced data, domain shift, and real-world constraints.
  • Deploy models to production, typically on the edge, instrument them for monitoring, and iterate with continuous integration and continuous delivery.
  • Write technical sections of proposals and white papers, help shape capture strategy, provide level-of-effort estimates, and present prototypes.
  • Present results to high-level Department of Defense and industry customers, author technical reports, and publish novel work in classified or unclassified settings when applicable.
  • Set technical direction, build a local engineering team, and ensure Department of Defense partners receive tangible mission impact from artificial intelligence and machine learning solutions.
Desired Qualifications
  • At least 8 years of experience designing, building, and deploying machine learning systems in production.
  • Prior Department of Defense project experience, including briefing senior stakeholders and navigating Department of Defense acquisition processes.
  • An active Secret or Top Secret/Sensitive Compartmented Information clearance and experience working in classified environments.
  • Experience managing or mentoring engineers and contributing to hiring and performance development.
  • Experience with proposal writing, including SBIR/STTR and requests for proposals, technical volume authorship, and live demonstrations.
  • Experience with multi-sensor fusion, signal processing, tracking, synthetic data generation, domain adaptation, or semi-supervised or active learning for data-limited environments.
  • Familiarity with edge hardware optimization and performance profiling.
  • Open-source contributions, publications, or patents in machine learning or computer vision.
  • An M.S. or Ph.D. degree.

About the company

CoVar develops customized machine learning and AI software for defense, healthcare, and industrial sectors. Its solutions are mission-critical systems built to operate reliably on high-stakes tasks, such as monitoring safety systems in drilling operations. The company starts by understanding a client’s unique challenges and then uses its team of technologists to apply AI and ML techniques to create tailored software and services. For example, its Adaptive Alarms software, developed with partners like Shell and Pason, uses advanced algorithms to prevent disasters on offshore and land drilling rigs, protecting workers and the environment. CoVar differentiates itself by delivering specialized, high-value, robust software rather than generic tools, emphasizing reliability and safety in demanding environments. The company’s goal is to be a trusted partner that helps large organizations improve performance, safety, and operational capabilities through customized AI solutions.

Company Size

11-50

Company Stage

N/A

Total Funding

N/A

Headquarters

McLean, Virginia

Founded

2011

Get referred to CoVar

See people who can refer or advise you

Simplify Jobs

Simplify's Take

What believers are saying

  • Warden 2026 validated BullsEye against live drones, feeding ATAK during a two-week Army demo.
  • March 2026 DARPA VectoredAI award reached $1.8 million, extending funded runway and validation.
  • June 2026 job postings for Army, ML, and signal roles signal active program expansion.

What critics are saying

  • CoVar still depends heavily on DARPA; contract loss would crush growth within 12 months.
  • 2026 hiring posts show lean cash and five open roles, signaling fragile financing.
  • BullsEye faces fast-moving counter-UAS rivals from Anduril and Dedrone with stronger sales machines.

What makes CoVar unique

  • BullsEye runs on soldiers' existing Android and Nett Warrior devices, avoiding new hardware.
  • CoVar’s DARPA work spans ASIMOV, PPADM, and VectoredAI, proving deep defense credibility.
  • Adaptive Alarms with Shell and Pason shows rare mission-critical AI across industrial rigs.

Help us improve and share your feedback! Did you find this helpful?

Benefits

Health Insurance

Dental Insurance

Vision Insurance

401(k) Company Match

401(k) Retirement Plan

Flexible Work Hours

Tuition Reimbursement

Company News

PR Newswire
Jun 9th, 2026
CoVar's BullsEye detects drones using soldiers' existing Android devices at US Army Warden 2026

CoVar demonstrated its BullsEye counter-drone capability at the US Army's Warden 2026 event, successfully detecting and tracking unmanned aircraft using standard Android devices and Nett Warrior systems that soldiers already carry. The technology requires no dedicated counter-drone hardware. During the two-week demonstration, BullsEye detected and localised small unmanned aircraft threats using only the devices' built-in RGB cameras, operating both as standalone units and networked systems. Networked devices triangulated to determine absolute geolocation, feeding target data directly into ATAK for a common operating picture. The system runs on commercial off-the-shelf Android devices using existing hardware including cameras, microphones and onboard computing. CoVar will continue developing BullsEye for potential deployment with soldiers.

PR Newswire
Apr 9th, 2026
CoVar wins DARPA contract to develop AI that adapts to human decision-makers' personalities

CoVar, a US Department of Defence AI specialist, has been awarded a Direct-to-Phase II SBIR contract by DARPA for its Predictive Psychological Architectures for Decision-Making programme. The contract supports development of VectoredAI, a dynamic human-AI teaming model that uses probabilistic key decision-maker attributes and latent semantic steering to align AI solutions with critical decision-makers' personalities. The PPADM programme aims to improve understanding of how individual characteristics influence decision-making behaviour, particularly in high-uncertainty, time-constrained scenarios where decision-makers disagree or have conflicting values. CoVar is partnering with Dr Jim Neidhoefer, combining CoVar's GEARS software with Neidhoefer's statistical methods for quantifying subjective factors. The award continues CoVar's work building trusted AI solutions for mission-critical defence systems.

The Bulletin
Apr 9th, 2026
CoVar awarded Direct-to-Phase II DARPA SBIR contract for Predictive Psychological Architectures for Decision-Making (PPADM).

CoVar awarded Direct-to-Phase II DARPA SBIR contract for Predictive Psychological Architectures for Decision-Making (PPADM). PR Newswire Today at 9:05am PDT MCLEAN, Va., April 9, 2026 /PRNewswire/ - CoVar, a small business specializing in advanced artificial intelligence and machine learning solutions for the U.S. Department of Defense, today announced it has been awarded a Direct-to-Phase II Small Business Innovation Research (SBIR) contract by the Defense Advanced Research Projects Agency (DARPA). The award supports a DARPA effort titled Predictive Psychological Architectures for Decision-Making (PPADM). The PPADM program aims to advance the scientific understanding of how individual decision-maker characteristics influence behavior, with the goal of improving both human and algorithmic decision-making, delegation, and trust. Focus will be placed on decision-making in difficult domains with high uncertainty, time and/or resource constrained scenarios, and circumstances in which decision-makers disagree or have conflicting values. CoVar will be responsible for developing VectoredAI, a dynamic human-AI teaming model that leverages probabilistic key decision-maker attributes (KDMAs) and latent semantic steering. VectoredAI is intended to provide a plug-in capability to help align AI solutions with personalities and considerations for critical decision makers. CoVar has teamed with Dr. Jim Neidhoefer to bring complimentary expertise in developing VectoredAI. CoVar's GEARS software, developed under DARPA's ASIMOV program, enables users to design world scenarios and evaluate dilemmas in terms of tensions and conflicts, supporting automated testing of both algorithmic and human decision making. Dr. Neidhoefer has developed statistical methods for quantifying and aggregating hard-to-measure subjective factors (like KDMAs) in addition to a methodology for aligning human and algorithmic decision making. "Barriers to AI adoption in mission critical and safety critical systems remains high due to lack of human trust," said Dr. Pete Torrione, CTO at CoVar. "Our proposed VectoredAI solution to PPADM will expand on what is already known about human decision making to identify key trust-building factors that reduce friction in human-AI interactions and humans' willingness to delegate difficult decisions to AI." The Direct-to-Phase II award represents CoVar's continued leadership in building trusted and responsible AI/ML solutions for the Department of Defense; ultimately leading to more rapid adoption of AI for mission critical systems. SOURCE CoVar This is a paid placement. For further inquiries, please contact PR Newswire directly.

PR Newswire
Dec 20th, 2024
Covar Awarded Darpa Contract To Define Ethical Standards For Future Autonomous Systems

MCLEAN, Va., Dec. 20, 2024 /PRNewswire/ -- CoVar, a leading developer of AI and machine learning solutions for the Department of Defense, has been awarded a multi-year contract with the Defense Advanced Research Projects Agency (DARPA) to support the Autonomy Standards and Ideals with Military Operational Values (ASIMOV) program. The goal of the program is to establish a framework that quantitatively measures the ethical readiness of autonomous systems, a crucial step as these systems gain widespread adoption across military and civilian sectors