Building firmware solutions for MCU platforms | Passionate about RTOS development, bare-metal programming, and edge AI
I specialize in developing embedded systems from hardware to software, with expertise in real-time operating systems, microcontroller programming, FPGA design, and deploying machine learning models on resource-constrained devices.
High-performance object detection pipeline combining TensorRT-optimized YOLO inference with low-latency Rust networking on NVIDIA Jetson Orin Nano. Features Unix socket IPC for inter-process communication and concurrent UDP/TCP streaming.
- Tech: TensorRT FP16, Rust (Tokio async), Python, CUDA, Unix Sockets
- Performance: 248.7 qps throughput, 4.6ms inference latency, <50ms end-to-end
- Architecture: Parallel pipeline (Camera β TensorRT β Rust β UDP/TCP streams)
Ported TensorFlow Lite Micro to ESP32 with custom porting layer, resolving CMSIS-DSP incompatibilities. Achieved production-grade inference performance through multi-core task scheduling and DMA optimization.
- Tech: ESP-IDF, TensorFlow Lite, CMSIS, CMake Build Systems, C/C++, FreeRTOS
- Performance: 55.1ms inference, 85KB RAM, 6.2 inferences/sec, 342KB total firmware
Implemented IEEE 754 single-precision floating-point arithmetic unit in Verilog. Hardware-accelerated computation for DSP applications.
- Tech: Verilog, Xilinx Vivado, FPGA
- Features: Addition, multiplication, division with pipelining
FSM-based stepper motor controller with configurable speed and direction control. Synthesized for FPGA deployment.
- Tech: Verilog, FPGA, PWM, FSM design
- Application: Robotics, CNC machines, automation
Automated test fixture for embedded hardware validation. Interfaces with multiple sensors and actuators for production testing.
- Tech: STM32, Python, I2C/SPI/UART protocols
- Features: Automated pass/fail criteria, data logging
Analytics dashboard for semiconductor manufacturing yield optimization. Real-time monitoring of hardware test results.
- Tech: Python, Pandas, Plotly, embedded data collection
- Application: Quality control, defect analysis
- π¬ Building TinyML models for ultra-low-power microcontrollers
- β‘ Optimizing real-time inference on edge devices (ESP32, Jetson)
- π οΈ Designing custom FPGA IP cores for signal processing
- π‘ Developing IoT sensor networks with LoRa/BLE connectivity
- π€ Integrating AI/ML into embedded robotics applications
I'm always excited to discuss embedded systems, hardware design, and edge AI projects!
- πΌ LinkedIn: anudeep-narala
- π§ Email: anudeepreddynarala@gmail.com
- π Open to collaboration on embedded systems and IoT projects