Embedded AI Software Engineer

We are seeking an Embedded AI Software Engineer with deep expertise in writing software for resource-constrained edge hardware. This role is critical to building optimized pipelines that leverage media encoders/decoders, hardware accelerators, and AI inference runtimes on platforms like NVIDIA Jetson, Hailo, and other edge AI SoCs.

You will be responsible for developing highly efficient, low-latency modules that run on embedded devices, involving deep integration with NVIDIA SDKs (Jetson Multimedia, DeepStream, TensorRT) and broader GStreamer pipelines.

Media Pipeline & AI Model Integration

• Implement hardware-accelerated video processing pipelines using GStreamer, V4L2, and custom media backends.
• Integrate AI inference engines using NVIDIA TensorRT, DeepStream SDK, or similar frameworks (ONNX Runtime, OpenVINO, etc.).
• Profile and optimize model loading, preprocessing, postprocessing, and buffer management for edge runtime.

System-Level Optimization

• Design software within strict memory, compute, and power budgets specific to edge hardware.
• Utilize multimedia capabilities (ISP, NVENC/NVDEC) and leverage DMA, zero-copy mechanisms where applicable.
• Implement fallback logic and error handling for edge cases in live deployment conditions.

Platform & Driver-Level Work

• Work closely with kernel modules, device drivers, and board support packages to tune performance.
• Collaborate with hardware and firmware teams to validate system integration.
• Contribute to device provisioning, model updates, and boot-up behavior for AI edge endpoints.

Educational Background:

• Bachelor’s or Master’s degree in Computer Engineering, Electronics, Embedded Systems, or related fields.

Professional Experience:

• 2–4 years of hands-on development for edge/embedded systems using C++ (mandatory).
• Demonstrated experience with NVIDIA Jetson or equivalent edge AI hardware platforms.

Technical Proficiency:

• Proficient in C++11/14/17 and multi-threaded programming.

• Strong understanding of video codecs, media IO pipelines, and encoder/decoder frameworks.

• Experience with GStreamer, V4L2, and multimedia buffer handling.

• Familiarity with TensorRT, DeepStream, CUDA, and NVIDIA’s multimedia APIs.

• Exposure to other runtimes like HailoRT, OpenVINO, or Coral Edge TPU SDK is a plus.

Bonus Points

• Familiarity with build systems (CMake, Bazel), cross-compilation, and Yocto.
• Understanding of AI model quantization, batching, and layer fusion for performance.
• Prior experience working with camera bring-up, video streaming, and inference on live feeds.
  

To apply, please send your resume and portfolio details to hire@condor-ai.com with “Application: Embedded AI Software Engineer” in the subject line.

Condor is an AI engineering company where we use artificial intelligence models to deploy solutions in the real world. Our core strength lies in Edge AI, combining custom hardware with optimized software for fast, reliable, on device intelligence. We work across smart cities, industrial automation, logistics, and security, with a team that brings over a decade of experience in AI, embedded systems, and enterprise grade solutions. We operate lean, think globally, and build for production from system design to scaled deployment.