Work Experience

A timeline of my professional journey.

Navy Federal Credit Union

2023 – PresentVienna, VA, USA

Senior Data Scientist

2023 – Present
2026 — Enterprise AI Strategy
  • TBD :)
2025 — AI/ML Center of Excellence
  • Built an automated LLM-as-a-Judge validation framework to detect incorrect classifications and missed topics, improving reliability and trust in large-scale LLM deployments.
  • Delivered GenAI-driven mortgage application and documentation insight discovery (leveraging the Insight Discovery Framework), producing structured complaint analysis, competitive comparisons, and executive-ready reporting.
  • Created a distilled sentiment analysis dataset and trained smaller models aligned with team-specific sentiment guidelines, enabling scalable cross-business adoption.
  • Designed and productionized a 6-month forward Engagement Tier predictive model, reducing ~1500 features to ~100 via feature selection and SHAP, with fully MLOps-ready training and scoring pipelines. The model was trained utilizing XGBoost and HyperOpt in Databricks.
  • Engineered graph-based member importance and relevancy features using PageRank-, TF-IDF-inspired, and custom metrics across transaction, referral, and product graphs.
  • Led LLM-based analysis of government shutdown impact, identifying affected members, extracting related topics, and quantifying sentiment shifts.
  • Standardized enterprise GenAI adoption by building reusable LLM templates, inference notebooks, and embedding pipelines adopted across multiple teams.
  • Delivered executive and cross-functional reports powered by the Insight Discovery Framework to support data-driven strategic decisions.
  • Trained and enabled multiple teams on LLM best practices, Databricks workflows, and Insight Discovery methodologies, presenting advanced generative AI techniques and optimization strategies across internal forums.
  • Re-architected and productionized PEGA-based marketing feature pipelines, engineering 11 feature-store-ready attributes and enabling scalable marketing model development.
  • Presented monthly at Databricks User Group, sharing internal NLP/GenAI solutions and advanced Databricks practices.
2024 — AI/ML Center of Excellence
  • Architected and productionized the enterprise Insight Discovery Framework (Map → Reduce → Classification), transforming large-scale call and chat data into structured intelligence used across departments. This framework can be used to surface product issues, member complaints, praise signals, competitive comparisons, and emerging themes, etc.
  • Enabled rapid post-launch feedback analysis for a major mobile app update (Omni V7), systematically surfacing complaints and praise and feeding insights directly to design and development teams for accelerated issue resolution.
  • Standardized LLM inference across the enterprise by building reusable GPU-optimized template notebooks (utilizing vLLM and LangChain), establishing the fastest production baseline.
  • Finetuned and deployed the organization's first 7B and 70B LLM models.
  • Designed NFCU's first internal assistant utilizing RAG technology in order to assist call representatives and other internal teams to automate their daily workflow.
  • Designed an automated LLM-as-a-Judge evaluation framework for RAG systems aligned with human scoring.
  • Delivered transcript-driven insight analyses (issues, benefits, competitive comparisons, etc.) for CD and credit card products that informed and influenced product revamp discussions.
  • Built a hybrid transformer + LLM praise detection pipeline to isolate high-impact praise and extract reasoning. Created the equivalent for complaint detection.
2023 — AI/ML Center of Excellence
  • Built and deployed the first-generation Voice of Member pipeline, transforming call and text data into structured sentiment, summaries, keyphrases, and taxonomy-driven insights. This was NFCU's first production NLP model.
  • Designed and productionized Topic Analysis and Intent Identification models using BERTopic and transformer-based architectures to better understand interaction drivers.
  • Delivered cross-functional ad-hoc analytics powered by NLP models, enabling data-driven decision-making across business units.

Data Scientist

2023 – 2023
2023 — AI/ML Center of Excellence
  • For detailed accomplishments during this period, see the 2023 section under the Senior Data Scientist role above.

Georgia Institute of Technology

2022 – 2023Atlanta, GA, USA

Graduate Teaching Assistant

2022 – 2023
2022 – 2023
  • GTA for the courses CS 7641-Machine Learning and CSE 6242-Data and Visual Analytics

General Electric

2022 – 2022Atlanta, GA, USA

Data Science Intern

2022
2022 — Data Science Intern
  • Created Deep Learning models based on the ANN, CNN and LSTM architectures using Tensorflow
  • Explored multiple model size reduction techniques
  • Deployed on an Arduino using Tensorflow Lite in order to perform inference on the edge

HelcoML Systems

2019 – 2021Athens, Attica, Greece

Data Scientist

2019 – 2021
2019 – 2021
  • Audio Deep Learning Model Training (TensorFlow, Keras)
  • Machine Learning Applications (XGBoost, Scikit-Learn)

National Centre for Scientific Research "Demokritos"

2018 – 2018Athens, Attica, Greece

Research Intern

2018
2018 — Institute of Nanoscience and Nanotechnology
  • Electrical characterization of memristive devices (memristors)
  • Analyzed the behavior of memristors during consecutive read and write tests
  • Executed retention measurements of memristors
  • Analyzed the behavior of memristors under various temperatures
  • Analyzed the analog behavior of memristors