RF & Edge AI Intern (Video Analytics)

What You'll Do

• ) Train, evaluate, and iterate drone‑detection models using video datasets and practical performance targets.
• Build video pipelines to handle preprocessing, inference, post‑processing, and event triggering (e.g., tracking, confidence scoring, alert logic).
• Deploy optimized models on constrained edge platforms, applying quantization or pruning where appropriate to meet latency, power, and memory limits.
• Design and execute benchmarking experiments to measure accuracy, false positives/negatives, robustness to lighting/weather/background clutter, and end‑to‑end latency.
• Maintain structured data and experiment tracking to ensure reproducibility (datasets, configs, metrics, model versions).
• Communicate technical findings through concise reports and demos, providing clear recommendations and next steps.
• Example Internship Deliverables (4–6 Months) A working edge demo for drone detection on video (live camera or recorded stream), including an alert/annotation overlay A measured benchmark report (precision/recall, false alarms, latency, compute footprint) and a proposed improvement plan A deployment-ready inference package (container/app/script) with documented setup and test procedure Requirements Current Masters or PhD student in Electrical/Electronic Engineering, Computer Engineering, Computer Science, or related field (enrolled throughout the internship) Solid ML foundations, including CNN-based vision models and evaluation metrics (precision/recall, ROC, confusion matrix) Hands-on experience with at least one ML stack (PyTorch preferred, TensorFlow acceptable) Experience with video/computer vision tooling (e.g., OpenCV, FFmpeg) and building practical pipelines Programming ability in Python; C/C++ is a plus for performance-critical edge work Strong technical communication: able to explain what you tried, what happened, and what it means Nice to Have Experience with edge runtimes and optimization (ONNX, TensorRT, TFLite, OpenVINO) Familiarity with embedded/Linux deployment and profiling (GPU/NPU acceleration, memory/latency profiling).
• Preferably with NVIDIA GPU environment.
• Exposure to detection/tracking architectures (e.g., YOLO-family, SSD, DETR, DeepSORT/ByteTrack) Experience with dataset curation/labelling strategies and handling class imbalance Understanding of real-world sensing constraints (camera optics, motion blur, range, occlusion) What We Offer Close mentorship from a technical team building real edge systems A defined project with milestones and an end-of-internship technical readout Exposure to the full lifecycle: data → model → deployment → benchmarking A collaborative team environment with strong learning culture For positions requiring access to technical data, Analog Devices, Inc. may have to obtain export licensing approval from the U.S.
• Department of Commerce - Bureau of Industry and Security and/or the U.S.
• Department of State - Directorate of Defense Trade Controls.
• As such, applicants for this position – except US Citizens, US Permanent Residents, and protected individuals as defined by 8 U.S.C. 1324b(a)(3) – may have to go through an export licensing review process.
• We foster a culture where everyone has an opportunity to succeed regardless of their race, color, religion, age, ancestry, national origin, social or ethnic origin, sex, sexual orientation, gender, gender identity, gender expression, marital status, pregnancy, parental status, disability, medical condition, genetic information, military or veteran status, union membership, and political affiliation, or any other legally protected group.
• Job Req Type: Internship/Cooperative Required Travel: No