I’m a Data Scientist based in Bengaluru, India. I thrive on turning raw data into actionable insights and building intelligent systems that make information accessible to everyone. With a strong foundation in deep learning, natural language processing and large language models (LLMs), I enjoy creating agentic AI assistants and data‑driven solutions for real‑world problems. I’m currently pursuing my M.Tech. in Computer Science & Engineering (Data Science) and hold a B.Tech. in the same field. My industry experience (2+ years) spans multimodal retrieval‑augmented generation, SQL agents, and scalable ETL pipelines.
At Philips, I work on multi‑agent AI systems and data engineering projects:
- Oral Healthcare Chatbot – Multimodal RAG AI: Developed a LangGraph‑powered, multimodal, multi‑agent system that retrieves insights from oral healthcare project archives across formats such as CSV, Excel, DOCX, PPTX and PDF.
- TextSense AI – SQL Agent: Built an agentic assistant that generates and executes SQL queries on Redshift to deliver natural‑language and tabular insights, enabling non‑technical teams to explore data.
- Patent Document Assistant: Currently developing an AI assistant that understands and summarizes complex patent documents using NLP techniques.
- Data Engineering: Designed and orchestrated robust ETL pipelines with Apache Airflow, implemented GDPR‑compliant data retention pipelines and architected data‑attestation workflows for security risk assessments.
I love to experiment with new ideas—here are some highlights from my repositories and portfolio:
| Repository | What it’s about | Highlights |
|---|---|---|
| data_analyst_agent | A Data Analyst Agent that leverages LangGraph and Streamlit to provide interactive analytical workflows. | Uses LangGraph’s agentic capabilities to query and visualize datasets in a Streamlit interface. |
| machineLearning | A curated collection of machine learning projects, covering tasks such as income prediction, object detection, sentiment analysis and stock prediction. | Showcases end‑to‑end notebooks and scripts across domains—from income and car price prediction to music genre classification. |
| genai_chatbot | A generative AI chatbot that answers queries about the CCMT counselling process using retrieval‑augmented generation. | Features a LLAMA‑based model with guardrails to reduce hallucinations and optimized inference using VLLM. |
| Practice_GenAI | A sandbox for exploring generative AI techniques and prompt engineering. | Contains exercises, model implementations and curated resources on GenAI and prompt design. |
| Python GUI Projects | A set of GUI applications built using Tkinter and PyQt5. | Includes a download manager, fantasy cricket game, ticket reservation system, speed tester, tic‑tac‑toe and age calculator. |
- Generative AI Counselling Assistant: Developed a retrieval‑augmented assistant to help students with counselling processes using FAISS for vector search and a guardrail system to mitigate hallucinations.
- Bird Species Identification: Designed an audio transformer model that classifies 105 bird species using scraped data from xeno‑canto and achieved 88 % accuracy.
- River‑Network Extraction: Built a CNN model to extract river networks from satellite images with data‑augmentation techniques for improved generalization.
| Category | Tools & Technologies |
|---|---|
| Languages | Python, C, SQL |
| Frameworks | LangGraph, LangChain, Apache Airflow, Streamlit |
| Libraries | TensorFlow, PyTorch, Transformers, scikit‑learn, Pandas, Psycopg2, Boto3, PySpark, Pytest |
| Databases | Amazon Redshift, PostgreSQL |
| Dev Tools | VS Code, Git/GitHub, Azure DevOps, CI/CD pipelines |
- Qualified GATE 2022 and topped the first year of B.Tech.
- Published a research article on Extraction of River Network from Satellite Image.
- Microsoft GenAI Hackathon Finalist.
I'm always excited to collaborate on challenging problems in data science, AI and software engineering. Feel free to reach out via:
- Email: [email protected]
- LinkedIn: animesh‑py
- GitHub: animesh1012
✨ Thank you for visiting! Let’s connect and build something amazing together.