DROPZONE ROBOTICS
Software Engineer (AI-Enabled Perception Systems)
DROPZONE ROBOTICS, Los Angeles, California, United States, 90079
Overview
We're looking for a hands-on AI and Embedded Software Engineer to help integrate real-time video perception capabilities for wearable and unmanned ground vehicle (UGV) edge systems designed for emergency medical scene awareness. You'll collaborate with a small, fast-moving team that includes experts in robotics, embedded systems, and tactical medicine. Your work will bridge AI model development, integration, embedded software development, and system-level optimization to help create a field-deployable system that works reliably in resource-constrained real-world environments with the potential to save lives. Responsibilities Develop and maintain real-time
perception processing pipelines
for video and multimodal sensor data (RGB, thermal, IMU, depth) Integrate, refine, and deploy standard AI/ML models (e.g., object detection, pose estimation, activity recognition) on embedded edge devices (e.g., NVIDIA Jetson, ARM SoCs). Implement
multi-object tracking
and
object re-identification
across frames and scenes Develop embedded software modules in C++/Python for low-latency sensor interfacing and synchronization. Design
structured logging
and
lightweight data storage
for detected entities/events (JSON, SQLite, ROS bag). Build tools to visualize, test, and debug perception outputs under varying lighting, motion, and occlusion conditions Profile and
optimize runtime performance
(e.g., TensorRT, ONNX Runtime, CUDA, multithreading). Build testing and
visualization tools
to validate robustness under varying conditions (lighting, occlusion, motion). Contribute to
system integration and reliability
in field-deployable, power- and compute-limited environments.
Requirements
Proficiency in
Python
and
C++
(for embedded and performance-critical code). Hands-on experience with
PyTorch
or
ONNX -based AI models. Strong foundation in
computer vision
(OpenCV, video processing, tracking). Experience working in
Linux-based embedded environments . Familiarity with
Git, command line tools, and build systems
(CMake, Make, CI/CD). Ability to debug, profile, and optimize software for real-time performance. Strong
attention to detail
and
curiosity
in learning new tools and workflows
Nice to Have
Experience with
NVIDIA Jetson ,
ARM , or other edge AI platforms. Knowledge of
TensorRT ,
CUDA , or hardware acceleration libraries. Familiarity with
ROS2
or other robotics middleware. Exposure to
sensor hardware integration
(cameras, IMUs, stereo, thermal). Background in
real-time systems, robotics, or embedded AI applications . Experience with
Jetson ,
TensorRT , or embedded GPU devices Familiarity with standardized models such as
MediaPipe ,
MMPose , or
Ultralytics YOLOv8+ See your work deployed in real-world life-saving search-and-rescue response Collaborate with domain experts from NASA, DoD, and frontline medics Fast-paced environment with flexibility, mentorship, and growth opportunities
Seniority
Entry level
Employment type
Full-time
Job function
Engineering and Information Technology
#J-18808-Ljbffr
We're looking for a hands-on AI and Embedded Software Engineer to help integrate real-time video perception capabilities for wearable and unmanned ground vehicle (UGV) edge systems designed for emergency medical scene awareness. You'll collaborate with a small, fast-moving team that includes experts in robotics, embedded systems, and tactical medicine. Your work will bridge AI model development, integration, embedded software development, and system-level optimization to help create a field-deployable system that works reliably in resource-constrained real-world environments with the potential to save lives. Responsibilities Develop and maintain real-time
perception processing pipelines
for video and multimodal sensor data (RGB, thermal, IMU, depth) Integrate, refine, and deploy standard AI/ML models (e.g., object detection, pose estimation, activity recognition) on embedded edge devices (e.g., NVIDIA Jetson, ARM SoCs). Implement
multi-object tracking
and
object re-identification
across frames and scenes Develop embedded software modules in C++/Python for low-latency sensor interfacing and synchronization. Design
structured logging
and
lightweight data storage
for detected entities/events (JSON, SQLite, ROS bag). Build tools to visualize, test, and debug perception outputs under varying lighting, motion, and occlusion conditions Profile and
optimize runtime performance
(e.g., TensorRT, ONNX Runtime, CUDA, multithreading). Build testing and
visualization tools
to validate robustness under varying conditions (lighting, occlusion, motion). Contribute to
system integration and reliability
in field-deployable, power- and compute-limited environments.
Requirements
Proficiency in
Python
and
C++
(for embedded and performance-critical code). Hands-on experience with
PyTorch
or
ONNX -based AI models. Strong foundation in
computer vision
(OpenCV, video processing, tracking). Experience working in
Linux-based embedded environments . Familiarity with
Git, command line tools, and build systems
(CMake, Make, CI/CD). Ability to debug, profile, and optimize software for real-time performance. Strong
attention to detail
and
curiosity
in learning new tools and workflows
Nice to Have
Experience with
NVIDIA Jetson ,
ARM , or other edge AI platforms. Knowledge of
TensorRT ,
CUDA , or hardware acceleration libraries. Familiarity with
ROS2
or other robotics middleware. Exposure to
sensor hardware integration
(cameras, IMUs, stereo, thermal). Background in
real-time systems, robotics, or embedded AI applications . Experience with
Jetson ,
TensorRT , or embedded GPU devices Familiarity with standardized models such as
MediaPipe ,
MMPose , or
Ultralytics YOLOv8+ See your work deployed in real-world life-saving search-and-rescue response Collaborate with domain experts from NASA, DoD, and frontline medics Fast-paced environment with flexibility, mentorship, and growth opportunities
Seniority
Entry level
Employment type
Full-time
Job function
Engineering and Information Technology
#J-18808-Ljbffr