ARKA

Machine Learning Engineer & Full Stack Developer

Kolkata, India
SCROLL

About & Skills

The Identity

I'm a dual degree graduate specializing in Machine Learning and Data Science with a strong foundation in full-stack development. My technical journey spans from developing deep learning models to building responsive web applications.

With experience in PyTorch, pandas, and the React ecosystem, I hunt down complex problems at the intersection of data science and software engineering. I'm particularly driven to create intuitive interfaces for complex systems.

Achievements

  • LeetCode 1500+ rating (Top 15% globally)
  • Won Code Relay 3.0 Hackathon organized by IIT Bhubaneswar
  • Achieved top 120 rank in Kaggle Sticker Sales Forecasting competition

Education

B.Sc. in Data Science and Programming

IIT Madras

2023 - 2027

CGPA: 9.0/10

B.Tech in Computer Science

Kalyani Government Engineering College

2023 - 2027

CGPA: 8.4/10

The Arsenal

Languages

JavaScriptPythonTypeScriptSQLHTML/CSSBash

Web Development

ReactFastAPIFlaskREST APIsResponsive Design

Tools

Git/GitHubDockerCI/CD (GitHub Actions)Linux/Unix

Databases

SQLDynamoDBNeo4jQdrant

Concepts

Scalable Web AppsMicroservicesAgile WorkflowsDebuggingTesting

The Cases

A selection of my recent missions spanning machine learning, data science, and web development.

Sticker Sales Forecasting

Sticker Sales Forecasting

  • Top 120 rank in Kaggle competition
  • Time series forecasting with ARIMA & Prophet
  • Feature engineering: sinusoidal & economic factors
  • EDA with 15+ visualization techniques
PythonScikit-learnPandasNumPyStreamlitMatplotlibSeabornPlotly
Real-time Sentiment Analysis Application

Real-time Sentiment Analysis Application

  • <10ms response time with ONNX optimization
  • React frontend + FastAPI backend
  • Transformer model inference optimization
  • Docker containerization + CI/CD pipeline
PythonFastAPIReactONNX RuntimeDockerCI/CD
RAG Institute: Hybrid Retrieval System

RAG Institute: Hybrid Retrieval System

  • Neo4j graph DB for entity relations
  • Vector search for semantic similarity
  • LLM integration for document Q&A
  • Hybrid retrieval architecture
PythonLangChainNeo4jLlamaIndexDocker