Skip to content
View adhishnanda's full-sized avatar
😄
😄

Block or report adhishnanda

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
adhishnanda/README.md

👋 Hi, I am Adhish Nanda

Data Analytics · Analytics Engineering · Applied ML & AI 📍 Berlin, Germany

Data professional with 2.5 years of industry experience in BI, ETL and reporting, currently completing an MSc in Data Science at the University of Europe for Applied Sciences in Berlin, graduating August 2026. I build complete systems rather than isolated notebooks: pipelines that run on a schedule, dashboards people actually use, and models that are evaluated properly rather than demoed once.

Currently: writing my MSc thesis on a governance first RAG system aligned with EU AI Act requirements, and applying for full time roles starting September 2026.


🛠️ Selected Projects

End to End dbt Analytics Stack dbt Core and BigQuery on 100k+ real e-commerce orders. 15 models across three layers, 94 automated data quality tests, GitHub Actions CI, live Looker Studio dashboard. → github.com/adhishnanda/ecommerce-dbt-stack

Personalised News Feed Ranking System LightGBM ranker, Redis online feature store, FastAPI serving layer, IPS and SNIPS counterfactual evaluation, 43 test suite. → Live demo: news-feed-ranker.streamlit.app → Code: github.com/adhishnanda/news-feed-ranking-system

Emotion Adaptive Multimodal System DistilBERT, CNN-BiLSTM and ResNet combined through late fusion, 54.1% accuracy on the MELD benchmark. → github.com/adhishnanda/emotion-adaptive-multimodal-cbt-assistant

Governance First RAG System (MSc thesis) Hybrid retrieval combining BM25, dense search and reranking, adaptive LLM routing, and evaluation through RAGAS and LLM as judge methods, aligned with EU AI Act and GDPR requirements.


💼 Background

2.5 years at NIIT Ltd as a Solutions Design Lead, building Power BI dashboards, SQL reporting workflows and ML case studies for enterprise training programmes, alongside a marketing analytics internship. Five peer reviewed publications in IEEE and Springer.

🎯 Open to

Data Analyst · BI Analyst · Data Engineer · Analytics Engineer · Junior Data Scientist · Junior ML or AI Engineer

Full time roles, based in Berlin and open to relocation across Germany, the Netherlands, Luxembourg, Ireland, Austria, Switzerland and the wider EU. EU Blue Card eligible upon a qualifying offer, no employer sponsorship required.

⚙️ Technical Skills

Languages & Querying

  • Python
  • SQL

Data Engineering & Analytics

  • dbt
  • BigQuery
  • Airflow
  • DuckDB
  • GitHub Actions

Business Intelligence

  • Power BI
  • Tableau

Machine Learning & AI

  • scikit-learn
  • LightGBM
  • PyTorch
  • FastAPI
  • Docker
  • Redis

📫 Contact

  • 🌐 Portfolio: adhishnanda.github.io
  • 💼 LinkedIn: linkedin.com/in/adhishnanda
  • 📧 Email: adhish.nanda@gmail.com

Pinned Loading

  1. ecommerce-dbt-stack ecommerce-dbt-stack Public

    Production-grade dbt + BigQuery analytics stack on 100k+ Olist e-commerce orders. Kimball star schema, 94 tests, CI/CD.

    Python

  2. news-feed-ranking-system news-feed-ranking-system Public

    Production-inspired personalized news feed ranking system with real-time content ingestion, interaction logging, feature engineering, ML ranking (Logistic Regression & LightGBM), and multi-objectiv…

    Python

  3. airflow-polling-intelligence-pipeline airflow-polling-intelligence-pipeline Public

    Production-style Apache Airflow ETL pipeline with feature engineering, batch ML prediction, and Streamlit dashboard for election polling analytics.

    Python

  4. emotion-adaptive-multimodal-cbt-assistant emotion-adaptive-multimodal-cbt-assistant Public

    Applied multimodal ML system combining text, audio, and visual emotion recognition with CBT-aligned response generation using late fusion.

    HTML

  5. motion-based-german-learning-app motion-based-german-learning-app Public

    Gesture-controlled, browser-based German learning app using computer vision and ML. Combines real-time MediaPipe gesture recognition with offline ML/DL experiments to inform robust system design.

    Jupyter Notebook

  6. healthcare-cost-prediction-analytics-course-project healthcare-cost-prediction-analytics-course-project Public

    Machine learning project for healthcare cost prediction, risk segmentation, and anomaly detection using Python, scikit-learn, and XGBoost.

    Jupyter Notebook