About Me
I am a passionate AI and data science enthusiast dedicated to harnessing the power of data to drive innovation and create meaningful solutions. With a deep-rooted curiosity for the world of AI and a love for extracting insights from data, I want to make a positive impact in the ever-evolving field. I also enjoy working in diverse and international teams and want to work with cutting-edge AI technologies that fuel my fervent passion and contribute towards a sustainable future.
Profile
Birthday
3 September 1996
Website
https://raviakash.github.io/
Master of Science
Information Technology
raaviakash03@gmail.com
Location
Rosdorf 37124, Göttingen, Germany
Research Interests
Deep Learning
Deep Reinforcement Learning
Computer Vision
Inference Optimization & Edge AI
Tech Stack I have worked with
Resume
Present
Junior Data Scientist
March 2024 - Present
Homiro Metrology GmbH, Göttingen, Germany
- Adapt and optimize HOMIRO AI for seamless integration into different AOI systems.
- Develop and implement Vision Transformers to ensure reliable detection of various types of defects on leather surfaces.
- Utilize TensorRT to optimize deep learning models for deployment in production lines, ensuring maximum efficiency and performance.
Education
M.Sc Information Technology
November 2020 - August 2023
Universität Stuttgart, Stuttgart, Germany
B.Tech Mechatronics - Robotics and Automation Engineering
July 2014 - April 2018
SRM Institute of Science and Technology, Chennai, India
Professional & Practical Experience
Master Thesis
February 2023 - August 2023
Bosch Center for Artificial Intelligence (BCAI), Renningen, Germany
- Study of different design choices enabling efficient learning in Model-Based Reinforcement Learning under sparse reward scenarios.
- Defined Integral probability metrics to assess the model state and reward distributions.
- Implemented ablation studies in continuous control environments, bench-marked novel ideas and compared against the baseline models.
- Leveraged MLflow to streamline the machine learning lifecycle, from experimentation and model tracking to deployment, enhancing efficiency and analysis capabilities.
Internship
June 2022 - December 2022
Bosch Research, Schwieberdingen, Germany
- Implemented automated Kafka data pipelines for eAxle time-series data in Hadoop, saving over 20 hours weekly by eliminating manual data retrieval.
- Created and optimized SQL queries using Pyspark in Hadoop edge node to extract, transform, and load data from various sources into a database.
- Combining measurement data with respective metadata in MongoDB for interactive visualizations and data analysis in Tableau. Digitalization Project
- Worked closely with DevOps teams in transitioning the data infrastructure to a containerized environment using Docker and further integrated with Azure Databricks.
Low-Voltage Electronics Engineer
September 2021 - August 2022
Mu-Zero Hyperloop, Karlsruhe Institute of Technology, Germany
- I was part of the team responsible for developing the custom hardware and software architecture of Electronic Control Units (ECUs) for the hyperloop pod's low voltage electronics
- ECU test-driven software development and generation of HAL using STMcube32 programmer.
- Developed Hardware/software test protocols for analyzing sensor signals through CAN messages using the CANoe tool.
- Conducted HIL test of the motor model and validated with test stand measurements.
Executive Engineer Maintenance and Capex Projects
September 2018 - September 2020
Carborundum Universal Limited, Chennai, India
- Implemented a PID control system for a pneumatic brake that maintains constant web ten- sion and achieved loss reduction by 50% - Six Sigma Project.
- Implemented Condition-Based Maintenance by estimating Remaining Useful Life using Ma- chine Learning algorithms. Reduced equipment cost by 12%.
- Industry 4.0 project - Automated data collection process and integrated with MySQL database for data analysis of machine performance, production data, and KPIs as part of IIOT implementation. Digitalization Project.
Projects
Check out some of my works!
- All
- Deep Learning
- Deep Reinforcement Learning
- Computer Vision
Testimonials
Carlos E. Luis
PhD Candidate @ Bosch Center for Artificial Intelligence
I supervised Ravi's Masters Thesis for six months. In this short period, Ravi was able to review the relevant literature, jump into an unfamiliar codebase, learn new tools for managing deep learning experiments and carry out several experimental studies. By the end of his MSc Thesis, Ravi acquired a deep understanding of state-of-the-art model-based reinforcement learning methods, which nicely complement his coursework on model-free RL. He was able to modify the code, design interesting new experiments, analyze results and discuss key insights. When faced with challenges, he was quick to propose new avenues to make progress. His Python skills allowed him to quickly test ideas, while also accounting for efficiency and readability of the code. I believe Ravi made great use of all the available tools to craft his MSc Thesis and learned lots of useful skills in that process. He's hardworking and excited to learn new things. He communicated clearly the project's progress and main challenges, which facilitated scientific discussions about his research topic. I'm confident that Ravi would be a great team player and I'm looking forward to see what the future holds for him.
Contact
Location:
Rosdorf 37124, Göttingen, Germany
Email:
raaviakash03@gmail.com