Work

Where I've worked and what I delivered.

  • Microsoft Corporation favicon
    Lead Applied Scientist / Lead Data Scientist
    May 2022 – Present 3 yrs 5 mo Berlin, Germany
    HealthcareGenAIEvaluationMultimodalRAGRAIEU
    • Leading applied research on multilingual clinical note generation for EU healthcare, with emphasis on compliance, safety, and adoption.
    • Designed evaluation pipelines (reference-free and reference-based) to assess hallucinations, omissions, and toxicity.
    • Led applied research on RAG systems for enterprise QA and retrieval tasks.
    • Served as Responsible AI Lead (EMEA), guiding 60+ RAI Champs and advising on compliance in sensitive domains.
    More details

    Role Overview:

    • Explored prompt strategies, fine-tuning, and architecture adaptation for clinical note generation.
    • Built agentic workflows integrating SFT and instruction-tuned models for steerable outputs in medical contexts.
    • Fine-tuned and evaluated LLMs (Phi-3-Vision, GPT-4, Mistral, LLaMA) across enterprise and multimodal retrieval tasks.
    • Collaborated on experimental pipelines for multimodal models (Phi-3-Vision, GPT-4o) in Healthcare and Autonomous Driving.
    • Developed scalable benchmarking setups for LLM search and retrieval performance.
    • Ensured compliance with EU AI Act and MDR regulations to enable trustworthy AI adoption in healthcare.
  • Continental AG favicon
    Senior Data Scientist (Product)
    Dec 2019 – Apr 2022 2 yrs 4 mo Hannover, Germany
    ForecastingMLOpsAWSSupply Chain
    • Developed end-to-end ML pipeline for demand forecasting AI platform in supply chain.
    • Reduced manual planning costs for 75+ planners by 10–15% using AWS cloud solutions.
    • Delivered 18-month rolling forecasts across 32+ EMEA markets.
    More details

    Role Overview:

    • Engineered features and evaluated ARIMA, Prophet, gradient boosting, and deep learning approaches.
    • Built reliable retraining and monitoring workflows to sustain accuracy in production.
    • Communicated results to stakeholders and managed platform user expectations.
  • Deloitte Consulting GmbH favicon
    Senior Data Scientist (Consultant)
    Sep 2017 – Nov 2019 2 yrs 2 mo Hannover, Germany
    ConsultingFinancial ServicesAutomotiveRetailForecastingNLPAnomaly Detection
    • Delivered ML solutions for the European Financial Authority, including credit risk prediction and anomaly detection on regulatory datasets.
    • Built an NLP-powered search engine for RfQ documents at a major automotive client.
    • Developed demand forecasting models for a leading German retailer.
    • Created internal Deloitte showcases, including fraud analytics, travel recommender, demand forecasting, and a reusable data science platform.
    More details

    Role Overview:
    European Financial Authority — Data Intelligence Service Center (2017–2018)

    • Developed models for credit risk forecasting.
    • Designed anomaly detection approaches for large-scale regulatory datasets.
    • Authored a user guide and educational material to scale adoption internally.

    Major Automotive Client (2018)

    • Built a domain-specific NLP search engine for document discovery and knowledge management.

    Major German Retailer (2019)

    • Designed daily demand forecasting models per store and product.
    • Consolidated production forecasts for suppliers into unified planning workflows.

    Deloitte Internal Initiatives (2017–2019)

    • Fraud Analytics Platform for detecting anomalies in financial datasets.
    • Travel recommender for employees.
    • Store-level demand forecasting tool.
    • Deloitte Data Science Platform for cross-use-case ML applications.
  • Jul 2016 – Jun 2017 11 mo Hannover, Germany
    Market ResearchAnomaly DetectionData StreamsData QualityMonitoring
    • Strengthened reliability of panel data by deploying a real-time anomaly detection system.
    • Accelerated predictive analytics projects with exploratory analysis and feature engineering.
    • Enabled faster decisions through dashboards that visualized and monitored data streams.
    More details

    Role Overview:

    • Focused on improving data quality and trust in large-scale panel research pipelines.
    • Designed anomaly detection approaches and validated them against historical datasets.
    • Partnered with internal analytics teams to roll out dashboards for quality tracking and reporting.
  • Basware Corporation favicon
    Technical Consultant
    Jun 2012 – Aug 2014 2 yrs 2 mo Chandigarh, India
    FinTechAutomationFinancial DataInternal Consulting
    • Designed and implemented automated workflows for financial data processing and e-invoicing.
    • Improved reliability of end-to-end data transmission across multiple financial data formats.
    • Contributed to projects enhancing transaction processing, digital invoicing, and client integrations.
    More details

    Role Overview:

    • Developed automation strategies for parsing and validating electronic invoice formats (MRT, EDIFACT, XML).
    • Enhanced stability of Basware’s financial transaction systems, including Virtual Printer and Transaction Server.
    • Delivered client-focused solutions ensuring smooth integration and accurate data exchange.
Education
  • Masters in Computer Science (Informatik)
    Leibniz Universität • Hannover, Germany
    Oct 2014 – Jun 2017
    • In collaboration with: TU Braunschweig, Uni Göttingen, TU Clausthal & KIT Karlsruhe.
    • Focus: Information Retrieval, Efficient Algorithms, Big Data, Data Mining, Simulation Engineering.
  • Bachelors in Computer Science (Informatik)
    UP Technical University • Lucknow, India
    Jul 2008 – Jun 2012
    • Focus: Engineering Mathematics, Data Structures & Algorithms, Data Mining & Warehousing, Distributed Systems.