JAYASHREE Parchuri

- I bring over 5 years of experience in Embedded Software Engineering, covering the entire Software Development Life Cycle for microprocessors, microcontrollers, SOCs, and IoT devices.
- My proficiency extends to various SDLC methodologies like Waterfall, Agile, and Test-Driven Development, complemented by strong coding skills in C, RTOS, and C++.
- I specialize in high-performance coding and IoT end-to-end application development, with a focus on real-time multithreading/multitasking in embedded applications.
- My experience includes developing drivers for RS232, I2C, SPI, and a deep understanding of high-speed display interfaces such as DP/eDP, HDMI, MIPI DSI, LVDS.
- I have hands-on expertise in Embedded Linux Kernel programming, board bring-ups, and bootloader, coupled with knowledge in configuring device tree, kernel Makefile, and the Yocto build process.
- Additionally, I possess extensive knowledge in Linux system calls, POSIX-based multithreading, Linux IPC, scheduling, and memory management.
- I am proficient in using various tools, including compilers, assemblers, simulators, and debuggers such as GCC, gdb, dbx, and Trace 32, to deliver high-quality embedded software solutions.

• Designed and implemented a predictive maintenance system for industrial IoT applications.
• Utilized machine learning algorithms (PyTorch, TensorFlow) to predict equipment failures before they occur.
• Integrated the system with industrial IoT sensors to collect real-time data for analysis.
• Worked on numpy, pandas, scikit-learn, OpenCV, and Tesseract to create and train machine learning models that transform audio and visual data into text data.
• Collaborated audio data using OpenAI's Whisper AI for transcription and diarization of real-time audio data into txt, TSV, and JSON formats.
• Utilized GCP's speech-to-text API to convert traditional video and audio datasets combined with scikit-learn for building machine learning models.
• Created a GUI using Tkinter to take the input of video data and save location in real time, displaying the result visually while also storing the files in the requested directory.