Skip to content
View Swish78's full-sized avatar
πŸ•Ί
Focusing
πŸ•Ί
Focusing

Block or report Swish78

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.

Please don't include any personal information such as legal names or email addresses. Maximum 100 characters, markdown supported. This note will be visible to only you.
Report abuse

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

Report abuse
Swish78/README.md

Yo, I'm Swayam! πŸ‘‹

πŸš€ Final-Year CSE Student | Machine Learning Enthusiast | Software Developer

I'm a final-year Computer Science Engineering student with a passion for Machine Learning, Deep Learning, and Software Development. I thrive on creating innovative solutions that make a positive impact. Always eager for new projects and collaborations in the tech realm.


🌟 Highlights of My Work:

  • πŸ¦™ Fine-tuned Llama 3 model on specialized datasets using Unsloth for cutting-edge NLP tasks.
  • πŸ” Implemented Retrieval-Augmented Generation (RAG) with Langchain and vector databases like Pinecone and Chroma.
  • ⚑ Developed real-time systems using WebSockets and Django Channels for seamless chat and notifications.
  • ☁️ Built and deployed ML models on AWS, including a credit risk prediction model using EC2 and S3.
  • 🌾 Fine-tuned YOLO for weed detection in agricultural fields to aid precision agriculture.
  • 🐳 Created isolated development environments with Docker, streamlining collaboration and development.

πŸš€ Recent Projects:

Celebro | FastAPI, React, Llama, ChromaDB (Dec 2024)

  • 🌐 Created an AI-powered search and query application with intelligent web search capabilities.
  • πŸ” Developed JWT-based user authentication with unique username and email verification.
  • πŸ€– Integrated Llama models for AI-enhanced response generation using Groq.
  • 🧠 Leveraged HuggingFace embeddings and ChromaDB for semantic search and vector storage.

StatsIQ - Data Analysis Platform | Llama, Flask, React, Neo4j, Scikit-learn (Nov 2024 – Present)

  • πŸ“Š Designed a SAAS platform for automated statistical analysis and machine learning model generation.
  • πŸ”’ Engineered secure file handling system supporting CSV/TSV uploads with data validation and sanitization.
  • πŸ” Constructed modular analysis pipeline incorporating descriptive statistics and correlation analysis.
  • πŸ“„ Generated automated LaTeX report system with customizable templates.
  • 🧬 Integrated Neo4j graph database and ChromaDB for vector operations.

VideoMaster | FastAPI, React, OpenCV, Tailwind CSS, AWS (Oct 2024)

  • πŸŽ₯ Built a browser-based video editing web application with multiple editing capabilities using OpenCV.
  • βœ‚οΈ Implemented video manipulation features including trimming, brightness adjustment, and speed control.
  • πŸ“½οΈ Supported multiple video export formats including MP4, AVI, and MOV.
  • ☁️ Deployed using AWS Lambda for backend processing and Amazon S3 for static website hosting.

🧠 Skills and Technologies:

  • Languages: Python C C++ R
  • Machine Learning: Scikit-learn NLP, Clustering
  • Deep Learning: TensorFlow CNN, LSTM, GANs, YOLO
  • Frameworks: HuggingFace Langchain Unsloth
  • Vector Databases: Pinecone ChromaDB
  • Web Development: Django Flask React FastAPI
  • Frontend: HTML Bootstrap Tailwind
  • Databases: MongoDB SQLite Oracle
  • Cloud Services: AWS (EC2, S3, RDS)
  • Tools: Docker Tableau

🀝 Collaboration Interests:

I'm always excited to collaborate on projects related to:

  • Machine Learning and Deep Learning applications
  • NLP and Data Science challenges
  • Full-stack Web Development
  • Real-time systems and cloud-based ML deployments

Feel free to reach out if you'd like to collaborate or discuss new ideas! Let's create something impactful together.


🌐 Connect with Me:

Profile Views


🎨 Random Dev Quote:

Quote


Pinned Loading

  1. Kaggle-Practice-Projects- Kaggle-Practice-Projects- Public

    Jupyter Notebook

  2. -WeedBuster-YOLOv8-Weeds-Detection- -WeedBuster-YOLOv8-Weeds-Detection- Public

    Jupyter Notebook

  3. CropCare-AI- CropCare-AI- Public

    Jupyter Notebook

  4. Celebro Celebro Public

    Tried creating clone of perplexity.

    TypeScript

  5. Jira-Lite Jira-Lite Public

    Jira Lite is a lightweight project management and task tracking application designed for internship programs. The application provides role-based dashboards for Project Managers, Team Leads, and Te…

    JavaScript

  6. VideoMaster VideoMaster Public

    VideoMaster is a sleek web app for effortless video editing, allowing users to trim, adjust brightness, change speed, and convert formats with a simple, intuitive interface and one-click processing.

    JavaScript