Allen Control Systems
Computer Vision/Machine Learning Intern, 2027
About this role
Join ACS's computer vision team as a summer intern working on real-time aerial threat detection and autonomous systems for deployed defense applications. You'll develop and optimize machine learning models for drone detection and tracking on embedded hardware alongside experienced robotics engineers.
What you'll do
- Develop and train computer vision models for real-time drone detection, tracking, and classification
- Build and maintain training data pipelines and datasets for model development
- Optimize models to run efficiently on embedded hardware like NVIDIA Jetson
- Conduct experiments in sensor fusion and tracking robustness under field conditions
- Collaborate directly with core engineering team on production-ready perception systems
- Debug and improve model performance in real-world deployment scenarios
What they're looking for
- PyTorch or TensorFlow
- Python programming
- Computer vision (OpenCV or similar)
- C++ (preferred)
- Real-time embedded systems optimization
- Machine learning model training and evaluation
- Data pipeline development
- Object detection and tracking algorithms
Benefits
- Competitive internship pay
- Equity package
- Health, dental, and vision insurance
- Paid time off
- Mentorship from experienced defense systems engineers
- Direct impact on deployed autonomous systems
Opens the application — the Jobs AI extension fills it for you. Set up autofill
Opens the official application on the employer’s site. No login required.
Allen Control Systems
Allen Control Systems builds autonomous defense systems, including gun turrets and drone-defense technology, using computer vision, control systems, and robotics. The company is hiring across software engineering, mechanical engineering, testing, and operations roles to support product development and scaled production.
View all jobs at Allen Control SystemsLikely interview questions
- Walk us through a computer vision project you've built—what dataset did you use, what architecture, and how did you measure success?
- How would you approach optimizing a PyTorch model to run in real-time on a Jetson device with strict latency requirements?