PhD Candidate · Hochschule München, Munich
Advancing Knowledge Discovery through Artificial Intelligence
AI Researcher · Information Retrieval Specialist · Patent Intelligence Expert · Builder of Domain-Specific AI Systems
I design intelligent systems that help organizations discover, retrieve, understand, and act on knowledge hidden inside patents, scientific literature, and large technical document collections — combining information retrieval, natural language processing, large language models, and explainable AI.
- 12
- Peer-reviewed publications
- 7+
- Years researching AI & NLP
- 2
- Countries — Germany & India
- 2026
- Expected PhD completion

Research Focus
Six pillars, one goal: turning unstructured knowledge into actionable insight
Patents are my proving ground — one of the densest, highest-stakes document collections in the world. The methods I build there extend naturally to scientific literature, legal documents, and enterprise knowledge bases.
IR · RANKING
Information Retrieval & Patent Search
Designing retrieval and ranking models that connect a search query to the most relevant prior art across millions of patent documents.
LLM · EVALUATION
LLMs for Prior-Art Search
Benchmarking large language models such as GPT and Gemini for prior-art search, and building evaluation frameworks for domain-specific retrieval tasks.
EXPLAINABLE AI
Explainable Search & Classification
Making AI-driven search and classification systems interpretable — highlighting the evidence behind every result for examiners and analysts.
EMBEDDINGS
Domain-Specific Language Models
Fine-tuning and quantizing transformer embeddings, including BERT-for-Patents, for efficient, domain-adapted retrieval.
TOPIC MODELS
Knowledge Discovery
Mapping large technical collections — from health-informatics patents to engineering literature — to surface emerging themes and structure.
TEXT MINING
Patent Analytics
Applying sentiment analysis, summarization, and text-mining techniques to extract structured insight from unstructured patent text.
Recognition
Achievements & milestones
Selected Publications
Recent work on patent retrieval & LLM evaluation
World Patent Information · Elsevier, 2025
Rethinking Patent Retrieval with Language Models: Toward Scalable and Efficient Search
ICAAI (ACM), 2025
Patent Retrieval with Few-Shot Fine-Tuning and Quantized Embeddings
PatentSemTech @ SIGIR, 2025
Enhancing Patent Retrieval Using Automated Patent Summarization
Interested in collaborating on AI-driven retrieval or patent intelligence?
Open to research partnerships, industry projects, and Indo-German initiatives in AI, NLP, and information retrieval.








