About Me
AI/ML Engineer with a knack for building smart, scalable, and full-stack AI systems.
Who I Am
I'm a Data Scientist and ML Engineer based in Berlin with over 5 years of experience building scalable machine learning pipelines and applying statistical models to solve real-world problems. Over the past year, I've also explored LLMs and RAG architectures, blending classic data science with cutting-edge AI.
My sweet spot is transforming complex data into practical, intelligent systems. I work across the stack—data ingestion, modeling, evaluation, and deployment—using tools like PyTorch, Python, Scikit-learn, and Hugging Face to create solutions that are both accurate and efficient.
I thrive on bridging ML, statistics, and AI, whether it's forecasting trends, classifying data, or fine-tuning models for retrieval and generation tasks. I’m passionate about turning raw data into actionable insights and intelligent applications.
5+
Years of Experience
1.5+
Years with LLMs
20+
Key Projects
M.Sc.
Data Science
Technical Skills & Proficiencies
A comprehensive overview of my technical expertise across various AI and development domains.
AI/ML Frameworks
LLM Tools
Backend & DevOps
Languages
Vector Databases
Professional Journey
My career path and key milestones in the field of AI and machine learning.
Data Scientist
Flix Mobility Tech GmbH – Berlin
Jan 2023 - Present
Enhanced tree-based demand prediction models for bus seat occupancy, improving accuracy through feature engineering and cross-team collaboration.
Data Scientist
Pixsy – Berlin
Jun 2021 – Jun 2022
Built an end-to-end ML pipeline for image similarity using Kafka & Spark. Developed a hybrid deep learning model (ResNet/Siamese) that improved matching accuracy by 22%.
M.Sc. in Data Science
Otto-von-Guericke Universität – Magdeburg
2019 – 2022
Key Courses: Machine Learning, Deep Learning, Learning Generative Models, Computer Vision, Recommenders, Data-Warehouse-Technologies.
ML Engineer
Accenture AI – Bengaluru
Nov 2018 – Aug 2019
Developed failure prediction models for trade transactions using Gaussian Mixture Models and optimized critical SQL queries, improving system response time by 38%.
Application Analyst
Accenture – Pune, India
Sep 2016 – Nov 2018
Built an email classification and ticket assignment model using Naïve Bayes and SVM (F1 score: 0.797). Performed EDA on retail sales data and optimized SQL queries, reducing response time by 38%.
B.Tech in Electronics & Communication
GGSIPU – Delhi, India
2012 – 2016
Key Subjects: Data Structures and Algorithms, C++, Software Engineering, Applied Mathematics and Statistics.
My Approach
The core principles that guide my work.
Full-Stack Mindset
Building end-to-end solutions, from robust data pipelines and backend logic to intuitive user-facing interfaces.
Pragmatic Innovation
Applying the right tool for the job, whether it's a classic ML model or a fine-tuned LLM, to deliver efficient and scalable results.
Continuous Learning
Staying hands-on with the latest tools and techniques in the fast-paced world of AI to build cutting-edge, optimized systems.