Welcome to My GitHub Page.

Explore my projects, skills, and experiences. Feel free to connect and collaborate!

• 🔭 I’m currently working on a GIT Based Model

• 🌱 I’m currently learning MLOps

• 💬 Ask me about LLMs , Recommendation Engines

• 📫 How to reach me [email protected]

• ⚡ Fun fact "Data scientists spend around 80% of their time cleaning and preparing data."

• Bachelor of Engineering (Honors) in Computer Science & Engineering

  • Assam Engineering College, Assam Science and Technology University (Aug 2019 - Jun 2023)
  • First Class, 81.4% (Top 5%)

• Intermediate (XII), Stream - Science

  • Maharishi Vidya Mandir Public School, CBSE (June 2017 – May 2019)
  • 87.6% (Grade: A)

• Matriculation (X)

  • SBOA Public School, CBSE (March 2016 – Feb 2017)
  • 10 CGPA (School Topper)

Da’Spatio Rhobotique Laboratory, Technology Incubation Centre Indian Institute of Technology Guwahati (IIT Guwahati) (Feb 2024 - Present)

• Contributing to the development of Convolutional Neural Networks (CNN) architectures for a research project focused on Image Captioning using Git-Based and Vision-Language Models.
• Utilizing PyTorch, Hugging Face Transformers and Datasets, and GPU acceleration.
• In-demand research skills: : Computer Vision, Neural Networks, Fine-Tuning, Exploratory Data Analysis, Feature detection and matching, Geometry.

Nestria (Byte) - Spryte Labs (Aug 2023 - Dec 2023)

• Gained extensive training in Data Analysis and Machine Learning, including EDA, Statistical concepts, Data Visualization, Feature Engineering, and implementation of machine learning pipelines.
• Conducted in-depth exploratory data analysis on large datasets and generated actionable insights for business decisions and strategy.
• Tech Stack: NumPy, Pandas, Matplotlib, Scikit-learn, Seaborn, Airflow.
• Repo Link : Upskilling-Analytics

• Empowering citizens with easy access to government policies through an AI chatbot that simplifies complex financial policies, offers personalized assistance, and enhances understanding of civic issues.
• Tech Stack: LLM, Fine-Tuning, Claude 3 - Opus, FastAPI, Next.js
• Live Link: Vercel + Render Deployment
• GitHub Link: Permisionless-AI-Chatbot

• Developed a 'Content Based' Recommendation System that suggests courses based on user's input of a selected course, utilizing techniques like Pre-processing, Count Vectorization, Stemming, and Cosine Similarity.
• Tech Stack: scikit-learn, FastAPI, Next.js
• Live Link: Vercel + Render Deployment
• GitHub Link: Educational-Course-RecommendationSys

• Developed a Flask app that fetches NEWS articles mentioning a given name as per its related sentiment, leveraging the Vader pre-trained NLP model, integrating Google custom search API, and implementing techniques like POS tagging and named entity recognition.
• Tech Stack: Vader, newspaper3k, NLTK, BeautifulSoup, Flask
• GitHub Link: Ethos-Hackathon-Final

Feel free to connect with me on LinkedIn or drop me an email. I'm always excited to collaborate on new projects and learn from others in the tech community.