I’m an AI engineer who builds production generative AI systems—large language models and agentic architectures designed for real, regulated workflows rather than notebook-only demonstrations.

My recent work focuses on turning complex financial and insurance documents into structured data for credit and insurance workflows. The hard part is not simply the model; it is making the output trustworthy. I therefore focus on grounding, deterministic validation, evaluation, and traceability so that outputs can be reviewed before they inform a decision.

Across NLP, computer vision, speech, and generative models, the throughline is the same: translating complex models into systems people can actually rely on. I work at the intersection of ML engineering, system design, and applied AI safety.

Experience

AI Engineer — additiv (May 2024 – Present · Zurich, Switzerland)

  • Build AI systems for document-heavy finance and insurance workflows, with a focus on structured extraction, retrieval, evaluation, and human review.
  • Design agent-based workflows using Azure-hosted language models, OCR, deterministic validation, and production APIs.
  • Work across system architecture, evaluation, monitoring, and deployment.

This description is intentionally limited to publicly shareable responsibilities; client names, internal project names, data, and implementation details are omitted.

Machine Learning Engineer — Validaitor (Apr 2023 – Mar 2024 · Karlsruhe, Germany)

  • Developed experimental pipelines for adversarial ML evaluation, including evasion, poisoning, and model-extraction scenarios.
  • Implemented watermarking and fingerprinting experiments to study model protection.
  • Evaluated LLM outputs for fairness, bias, and toxicity.

Research Assistant — Institut für Parallele und Verteilte Systeme, University of Stuttgart (May 2022 – Oct 2022 · Stuttgart, Germany)

  • Researched spatio-temporal word embeddings that incorporate temporal and geographic context.
  • Built Python-based embedding pipelines and evaluation frameworks to analyse temporal semantic shifts and regional language variation.
  • Explored downstream NLP tasks including time-sensitive text analysis and regional usage detection.

Machine Learning Engineer — Quantiphi (Jan 2021 – Oct 2021 · Bengaluru, India)

  • Worked on multimodal emotion recognition combining text and speech signals.
  • Built an unsupervised speaker diarization pipeline using voice activity detection, speaker representations, and clustering.

Machine Learning Engineer — Scanta Inc. (Mar 2019 – Mar 2020 · Gurgaon, India)

  • Coordinated a small team of ML engineers across experimentation and deployment.
  • Designed and deployed NLP pipelines on cloud infrastructure for text processing and language transformation.
  • Worked on data augmentation, language correction, and style-transfer models, evaluated on internal datasets.
  • Collaborated with product and leadership stakeholders to translate requirements into deployable ML systems.

Machine Learning Engineer — Mobile Programming LLC. (Jul 2018 – Dec 2018 · Gurgaon, India)

  • Built an attention-based encoder–decoder prototype for domain-specific machine translation.
  • Developed a BiLSTM-CRF prototype for named entity recognition in pharmacological and medical text.

Education

M.S. Computational Linguistics — University of Stuttgart

B.Tech. Computer Science — Dr. A.P.J. Abdul Kalam Technical University

Skills

Languages & Frameworks: Python, C++, PyTorch, TensorFlow, scikit-learn

LLMs & generative AI: LangChain, Hugging Face Transformers, Azure OpenAI, RAG pipelines, agentic AI, PEFT / LoRA

Cloud & Infrastructure: Azure (AI Services, Functions, Blob Storage), AWS, Google Cloud, Docker

ML Domains: NLP, Speech Processing, Computer Vision, OCR, Adversarial ML, Diffusion Models

Other: Knowledge Engineering, Prompt Engineering, Model Evaluation & Safety

Public work

Interests

Table Tennis, Chess, Financial Markets, Quantum Physics