● Design and Implement Micro-Services architecture. Use gRPC, Redis Pub/Sub to communicate.● Encourage good coding practices across teams.● Wrote containerized services for better deployment experience using dockers.● Implement CI/CD of services for automated build.● Design and develop software pipelines for low latency and high throughput.● Use intel and nvidia hardware encoders and decoders for image processing.● Use VA-API with GStreamer pipeline to write h264 encoded video with low cpu consumption and latency.● Wrote cuda kernals for data conversion on gpu to speedup cuda memcpy from host to device and increase FPS of inference engine.

● Implement Agile methodologies in software development and follow Test Driven Developmentpractice.● Deploy containerized system on client sites using docker-compose.● Wrote multithreaded C++ service to run TensorRT and Intel OpenVino Inference Engines withsupport of multiple network architectures.● Deploy systems that use various machine learning model on mobile embedded systems (likeNVIDIA® Jetson AGX XavierTM and NVIDIA® JetsonTM TX2).● Development on aarch64 systems.● Use Google Test for Unit testing of functions and Google Benchmarking for benchmarking offunctions.● Relevant programming languages/environments: Python3, C/C++, Bash