Final-year CS (Data Science) student building toward data engineering. I like taking messy, multi-source data and turning it into pipelines that actually hold up. Comfortable across the stack: ingestion, modeling, and the dashboards on top.
- Entity resolution system unifying ASHA, ANM, PHC, and Anganwadi community health records
- RapidFuzz fuzzy matching with Kannada transliteration handling; NFHS-5-validated synthetic data
- Final-year major project — KSCST Student Project Programme submission
- Built LightGBM model (93.5% weighted F1-score) outperforming RF & Logistic Regression
- Engineered temporal + weather-based features; handled imbalance with SMOTE
- Deployed Streamlit app with real-time predictions & SHAP explainability 👉 https://github.com/flimsycelery/Surge-Fare-Predictor
- Multilingual YouTube comment analysis using Groq + Llama 3.3 70B and BERTopic, across 20+ languages
- Applied text processing, embeddings, and classification 👉 https://github.com/flimsycelery/Comment-Culture-Analyzer
- Built semantic skill-matching using Sentence Transformers
- Designed NLP-based matchmaking algorithm using cosine similarity
- Backend with Flask, MongoDB, and real-time communication 👉 https://github.com/flimsycelery/SkillEx
- RESTful backend for authentication, products, and orders
- Designed scalable API structure with clean architecture 👉 https://github.com/flimsycelery/Ecommerce-API
- Microsoft DP-600 (Fabric Analyst)
- Microsoft DP-700 (Fabric Data Engineer)
- Microsoft DP-900 (Azure Data Fundamentals)
- Microsoft DP-750 (Azure Databricks)
- Oracle OCI 2025 Generative AI Professional
- Oracle OCI 2025 AI Foundations Associate
