Tip: "Save as PDF" in the print dialog for a clean one-pager.
Gulab Sana Parveen
AI/ML Engineer · LLMs, RAG, and secure AI systems
Mannheim, Germany · gulabsanaparveen@gmail.com
Summary
AI/ML engineer with expertise in software engineering, LLMs, and secure AI systems. Skilled in building scalable, production-ready solutions across retrieval-augmented generation, NLP, and computer vision, focused on solving real-world problems with cutting-edge technologies.
Experience
Data Scientist · SAP · Germany
Oct 2024 – Present- Designed and implemented a production-grade Retrieval-Augmented Generation (RAG) framework for enterprise case management and the SAP Help Portal, leveraging LLMs (Llama, Mistral, GPT-based models).
- Built and deployed a prompt framework for LLM integration across SAP AI services: ISM case summarization & root-cause analysis, outbreak detection for major-case narratives, and AI-based component prediction.
- Developed a dynamic, structured prompt-template system for varied document types (SAP Notes, KBAs, Community Articles) for consistent LLM reasoning and output quality.
- Led an Auto Prompt Tuning (APT) initiative: prompt-variation testing (grid/random search) and reinforcement learning (PPO/bandit) to optimise document-selection accuracy.
- Engineered secure prompt handling (fixed vs. user-modifiable segregation) to prevent prompt injection, and reasoning chains for explainability.
- Managed model lifecycle and deployment with Kubeflow and CI/CD, integrating security (Mend) and code-quality (Sonar) gates.
Machine Learning Engineer · SAP · Germany
Oct 2023 – Sep 2024- Ran benchmarking, fluctuation analysis, and root-cause investigations on ML recommendation systems and ISM/Coveo search pipelines.
- Designed a scalable human-feedback system (API + PostgreSQL) for ISM RAG, with binary, categorical, and fine-grained feedback aligned with Perplexity, GPT, and Bing.
- Led data analysis, visualization, and dataset creation over large-scale customer-incident data to improve AI recommendations and search accuracy.
Research Assistant · DFKI · Germany
Sep 2022 – Sep 2023- Worked on computer-vision OCR (PaddleOCR, EasyOCR, PyTesseract) and NER for the Trusted Research Environment (OpenSafely project).
- Contributed to a Secure ML Architecture for analysing sensitive electronic health records (EHR).
Senior Software Engineer · EPAM Systems · India
Jan 2018 – Feb 2022- Built smart applications for Mastercard GDPR with OCR text extraction and NER entity detection.
- Built an email classifier (Random Forest) to automate business operations.
- Full-stack development (Java, Spring, Angular) with a microservice architecture for a multinational financial client.
Product Development Intern · Adobe Systems · India
Early career- Performance enhancement for the resource-critical Live Edit Service (LES) in Java.
Education
M.Sc. Computer Science · RPTU Kaiserslautern
Apr 2022 – Nov 2024Specialisation: Intelligent Systems, Artificial Intelligence & Data Analytics.
Selected projects
Skills
- Languages
- Python (Pandas, NumPy, SciPy), Java, Julia, C
- AI / ML
- ML, Deep Learning, NLP, LLMs, RAG, Computer Vision, NER, OCR, Secure ML
- Frameworks
- PyTorch, TensorFlow, Hugging Face Transformers, spaCy, BERT, FastAPI, Django, Spring, Angular
- Cloud / MLOps
- AWS, GCP, Azure, PCF, Kubernetes, Kubeflow, Docker, Jenkins, CI/CD, MLOps
- Data
- PySpark, Data Modeling, PostgreSQL, MongoDB, SQL, OracleDB, ArangoDB, Power BI