About this role
Engineers who thrive in rapid, hands-on environments and want to close the gap between design and deployment will find a strong fit at Neros Technologies. The company builds high-performance unmanned systems for defense applications, operating with an urgent startup pace focused on getting capabilities directly to warfighters.
As the Perception Lead, this person will direct the perception stack across all vehicle platforms, managing detection, recognition, and image-space tracking from sensors to state estimation. Responsibilities include guiding a small engineering team while remaining hands-on with the code, establishing technical roadmaps, and overseeing the transition from classical computer vision to learned models as flight data warrants.
Candidates must possess extensive background in production computer vision and embedded deployment, bringing a readiness to handle performance challenges under degraded conditions like low light and thermal imaging. This position exists to ensure reliable target tracking in complex, real-world outdoor environments where precision matters.
What you will do
- Manage the complete architecture and delivery of the perception stack across all unmanned vehicle platforms with a unified codebase.
- Direct and mentor the perception engineering team while continuing to write production code directly.
- Deliver detection, recognition, and image-space tracking capabilities that maintain locks through clutter, glare, and low-light scenarios.
- Establish perception output contracts and coordinate detection operating points alongside state estimation owners.
- Transition systems from classical computer vision to learned detection and classification based on flight data performance.
- Deploy perception software onto constrained onboard compute hardware while balancing latency and accuracy constraints.
What you need
- Six or more years of experience building production computer vision or perception software.
- Minimum of two years guiding a team or technical program as an engineering manager or tech lead.
- Proven track record of shipping perception software onto physical systems operating in uncontrolled outdoor environments.
- Expert proficiency in both C++ and Python development for real-time or resource-constrained settings.
- Deep knowledge spanning classical tracking methods as well as modern detection and classification networks.
- Hands-on experience deploying neural networks to embedded accelerators such as NVIDIA Jetson, TensorRT, or equivalent hardware.
- Working familiarity with electro-optical and thermal imaging, camera calibration, and small-object detection at long range.
Nice to have
- Experience selecting and integrating thermal and LWIR sensors along with procurement constraints.
- Experience building or operating fleet data engines for large-scale log capture, indexing, search, and labeling.
- Working familiarity with state estimation and sensor fusion, including visual-inertial odometry or Kalman filtering.
- Experience participating in flight test or field test operations, range procedures, and log recovery.
Benefits
- Salary range of $201,000 to $281,000 USD
- Highly competitive equity grants
Skills and tech stack
Questions candidates ask
What is the salary range for this role?
The salary range for this role is an estimate between $201,000 and $281,000 USD based on a wide range of compensation factors. Actual salary may vary based on work experience, education, training, critical skills, and business considerations.
What components are included in the total compensation package?
The estimate is inclusive of base salary only. Highly competitive equity grants are also considered part of the total compensation package.
What kind of systems will I be working on?
You will own the perception stack across all Neros unmanned aircraft, covering detection, recognition, and image-space target tracking. The work involves deploying software to constrained onboard compute and physical systems operating in uncontrolled outdoor environments.
What technical background is required for this position?
You need six or more years developing production computer vision or perception software, including expert proficiency in C++ and Python. The role also requires depth in both classical and learned perception alongside experience deploying neural networks to embedded accelerators.
The application is on the official Neros Technologies careers page.
Apply on company site