About Me
I am a Machine Learning Engineer deeply passionate about artificial intelligence, computer vision, and large language models (LLMs). Originally from India, I began my academic journey with a Bachelor of Science in Computer Science, where I developed a strong foundation in programming and technology. This foundation led me to pursue a Master’s degree in Data Science and Artificial Intelligence at Saarland University in Germany.
My expertise lies in computer vision, where I utilize advanced techniques such as the YOLOv7 model. One of my notable projects is an innovative application for gym environments that offers real-time posture correction feedback. This project exemplifies my ability to merge technical skills with practical applications to enhance health and fitness routines.
In the domain of large language models (LLMs), I have developed advanced systems that assist in enhancing educational experiences. A significant project in this area involved creating an LLM-based chatbot for Universität Leipzig, designed to help students access course materials and manage exam registrations seamlessly. By leveraging advanced NLP techniques, I ensured that the chatbot could deliver contextually accurate responses, significantly improving the efficiency of academic administrative processes. This work highlights my ability to apply cutting-edge AI techniques to solve real-world problems in the education sector.
In addition to my strong background in computer vision and machine learning, I am also well-versed in ML operations. I understand the importance of deploying, monitoring, and maintaining machine learning models in production environments, ensuring they deliver reliable and efficient performance.
My comprehensive educational background and hands-on experience equip me to continuously explore new technological advancements. I am eager to contribute meaningful solutions to the AI landscape, driven by creativity, technical expertise, and a deep understanding of user needs. I am particularly interested in opportunities where I can combine my skills in computer vision, LLMs, and ML operations to drive innovation and operational excellence.
Professional Experience
Junior Researcher - Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI)
January 2023 - Present
Lead Engineer, FitSight: Personalized Recommendation Engine
Berlin, Germany
- Led the development of a real-time gym activity recognition system using YOLOv7, focusing on Machine Learning (ML), AI, and Python programming for data analysis, ML, and back-end development, alongside HTML, CSS, and JavaScript for front-end work.
- Deployed the system on Kubernetes with Kafka for real-time data processing. Implemented CI/CD pipelines and GitHub Actions for automated builds and deployments.
- Published research at major conferences. Maintained the project on GitHub: DFKI-fitsight.
- Skilled in YOLOv7, TensorFlow, PyTorch, with strong Python expertise. Proficient in Kubernetes, Kafka, Git, CI/CD pipelines, and GitHub Actions. Used Pandas, NumPy, and Matplotlib for data analysis and visualization.
- Live application available at: milki-psy.dbis.rwth-aachen.de/client.
- Contact: Milos Kravcik - Milos.Kravcik@dfki.de
Testing Manager - CTC Advanced Gmbh
June 2022 - December 2022
Smart Card Technologies
Saarbrucken, Germany
- Enhanced mobile authentication security by applying machine learning to biometric systems, significantly refining face and fingerprint recognition accuracy and robustness to prevent unauthorized access.
- Led the validation of advanced models to distinguish genuine users from spoofing attacks, greatly elevating mobile phone security and safeguarding sensitive data against sophisticated threats.
- Proficient in implementing ISO/IEC 7816 and ISO/IEC 14443 standards for smart card technology, ensuring compliance with global security protocols and enhancing the reliability of biometric authentication systems.
- Conducted risk assessments and implemented cryptographic algorithms (AES, DES, RSA) and protocols, identifying vulnerabilities in smart card systems, improving product performance, and strengthening defenses against potential security breaches.
Research Assistant - CISPA Helmholtz Center for Information Security (CISPA)
January 2022 - June 2022
Sensor Fusion for Autonomous Vehicles
Saarbrucken, Germany
- Conducted research on multi-sensor fusion to enhance the security of autonomous vehicles.
- Integrated 5G, IMU, and GPS sensors to detect and mitigate real-time attacks through anomaly detection.
- Developed and tested GPS spoofing attacks, using 5G and IMU data to identify and counteract these threats.
- Employed Google Maps API and LGSVL Simulator to collect and analyze data, validating the effectiveness of the fusion approach in ensuring safe and secure autonomous driving.
- Contact: Mridula Singh - Singh@cispa.de
Graduate Engineer Trainee - Xcelerator - Learn By Doing
January 2020 - April 2020
Bengaluru, India
Education
Master of Science (M.Sc) - Saarland University
10/2020 – present | Saarbrücken, Germany
Specialization: Data Science and Artificial Intelligence
Bachelor of Science (B.Sc) - CMR University
06/2016 – 06/2020 | Bengaluru, India
Specialization: Computer Science Engineering (CSE)
Projects
-
LLM Chatbot for Universität Leipzig | Python, NLP, LLM
June 2024 - present
Developed a Large Language Model (LLM) chatbot designed to assist Leipzig University students in accessing course materials and managing exam registrations. This project was aimed at enhancing student self-directed learning and streamlining administrative processes. By employing advanced Natural Language Processing (NLP) techniques, the chatbot was able to deliver contextually accurate responses with an accuracy rate of 84.78%. The system was seamlessly integrated with the university's existing infrastructure, providing students with an efficient and reliable tool for their academic needs.
-
Augmented Intelligence in Tutoring Systems & Tracking & Feedback Engine for Personalized Fitness Training | Yolov7, Kubernetes
January 2023 – June 2024
Engineered a scalable real-time human pose tracking system leveraging YOLOv7, specifically designed for fitness training applications. This project combined efforts from the "Augmented Intelligence in Tutoring Systems" and "Tracking & Feedback Engine for Personalized Fitness Training" initiatives. The system was developed to enhance self-directed learning by providing immediate posture correction and feedback during exercises. Advanced computer vision techniques and deep learning models were integrated to optimize the system for deployment in user-centric health applications. Emphasis was placed on achieving high model accuracy, reducing latency, and ensuring seamless integration with AI-driven feedback systems.
-
Multi Sensor Fusion for Autonomous Vehicles | LGSVL, Google Maps API
January 2021 – June 2022
Developed a comprehensive sensor fusion system for autonomous vehicles, aimed at enhancing the detection and mitigation of real-time attacks. The system integrated 5G, IMU, and GPS sensors to improve the safety and security of autonomous driving by detecting anomalies. Additionally, the project involved developing and testing GPS spoofing attacks, which were validated using Google Maps API and the LGSVL Simulator. The fusion approach proved effective in simulating and counteracting real-world threats, ensuring that the autonomous vehicles could operate safely under various conditions.
Publications
-
FitSight: Tracking and Feedback Engine for Personalized Fitness Training
Authors: Hitesh Kotte, Florian Daiber, Milos Kravcik, and Nghia Duong Trung
In Proceedings of the 32nd ACM Conference on User Modeling, Adaptation and Personalization (UMAP 2024), July 1–4, 2024, Cagliari, Italy. ACM, New York, NY, USA,
-
IMPECT-POSE: A Complete Front-end and Back-end Architecture for Pose Tracking and Feedback
Authors: Abhishek Samanta, Hitesh Kotte, Patrick Handwerk, Khaleel Asyraaf Mat Sanusi, Mai Geisen, Milos Kravcik, and Nghia Duong Trung
In Adjunct Proceedings of the 32nd ACM Conference on User Modeling, Adaptation and Personalization (UMAP Adjunct 2024), July 1–4, 2024, Cagliari, Italy. ACM, New York, NY, USA
-
Augmented Intelligence in Tutoring Systems: A Case Study in Real-Time Pose Tracking to Enhance the Self-learning of Fitness Exercises
Authors: Nghia Duong Trung, Hitesh Kotte and Milos Kravcik
18th European Conference on Technology Enhanced Learning (ECTEL2023), September 04–08, 2023, Aveiro, Portugal
-
Real-Time Posture Correction in Gym Exercises: A Computer Vision-Based Approach for Performance Analysis, Error Classification, and Feedback
Authors: Hitesh Kotte, Milos Kravcik, and Nghia Duong Trung
MILeS’23: 18th European Conference on Technology Enhanced Learning (ECTEL2023), 3rd International Workshop on Multimodal Immersive Learning Systems, September 04–08, 2023, Aveiro, Portugal.
-
Augmented Intelligence in Tutoring Systems: Real-Time Pose Tracking to Enhance the Self-learning of Fitness Exercises
Master Thesis - Saarland University (DSAI Dept.)
Author: Hitesh Kotte
Supervisors: Dr. Florian Daiber and Dr. Milos Kravcik
Reviewers: Prof. Dr. Antonio Krüger and Dr. Nghia Duong Trung
Contact Information
Personal : hiteshkotte@gmail.com
Work : hitesh.kotte@dfki.de
LinkedIn : linkedin.com/in/hiteshkotte
GitHub : github.com/hiteshkotte
Germany : +49 1606390778
WhatsApp : Send me a message
Date of Birth : 27/06/1998
Thank you for visiting my profile. I look forward to connecting with you!