{
  "name": "Shamik Basu",
  "headline": "AI Enabled Data Scientist",
  "summary": "Data scientist with 2+ years shipping production machine learning pipelines, LLM systems, and analytics workflows with measurable business impact. Cut operational costs 72% on a production RAG system processing 5M+ records/month and improved fraud identification accuracy 27% while reducing false positives. Strong in Python, SQL, statistics, experimentation, machine learning, and LLM/agentic systems.",
  "location": "Los Angeles, CA",
  "email": "shamik1900@gmail.com",
  "phone": "+1-213-255-6219",
  "website": "https://www.shamik-basu.com/",
  "interests": ["Table Tennis", "Football (Soccer)", "Piano", "Horror Movies", "Anime", "Sports"],
  "links": {
    "linkedin": "https://linkedin.com/in/shamikofficial/",
    "github": "https://github.com/ShamikOfficial",
    "devpost": "https://devpost.com/shamik1900",
    "website": "https://www.shamik-basu.com/",
    "resume": "assets/documents/resume.pdf"
  },
  "stats": [
    { "label": "Years production ML", "value": "2+" },
    { "label": "MS Data Science", "value": "USC" },
    { "label": "Hackathon wins", "value": "1st place" }
  ],
  "highlights": [
    {
      "header": "1st Place - UCLA SAIRS Hackathon",
      "blurb": "Won the sustainability track with EcoMate-AI, a multimodal agentic platform built in 24 hours.",
      "date": "2025",
      "link": "projects/ecomate-ai.html"
    },
    {
      "header": "Shipped Trend Pilot",
      "blurb": "Agentic AI for creator teams (FastAPI + Next.js + GPT-4o VLM) that catches trends, generates scripts, and assigns the work.",
      "date": "2026",
      "link": "projects/agentic-trend-orchestrator.html"
    },
    {
      "header": "Vice President - GRIDS at USC",
      "blurb": "Leading a 250-member graduate data science community with workshops, ideathons, and projects.",
      "date": "2025-Present",
      "link": "https://linkedin.com/in/shamikofficial/"
    }
  ],
  "experience": [
    {
      "role": "Data Science Associate Intern (Part-Time)",
      "company": "Liquidity AI Capital",
      "location": "Los Angeles, CA",
      "dates": "Apr 2026 - May 2026",
      "summary": "Deployed production SLM inference in Docker, fine-tuning open-source models end-to-end (tokenization + LoRA adapters) for optimized, domain-specific inference on financial data."
    },
    {
      "role": "Data Science Associate Intern (Part-Time)",
      "company": "KCC Capital Partners",
      "location": "Los Angeles, CA",
      "dates": "Jan 2026 - Mar 2026",
      "summary": "Automated M&A target sourcing with an agentic LangChain pipeline (Python) chaining LLM entity classification, web scraping, and structured extraction to identify and rank strategic buyers."
    },
    {
      "role": "Data Scientist",
      "company": "Bajaj Finserv Health",
      "location": "Pune, India",
      "dates": "Nov 2023 - Dec 2024",
      "summary": "Architected a production RAG pipeline (LangChain + GPT-3.5 Turbo + REST APIs) processing 5M+ medical records/month - cutting operational costs 72% at 92% extraction accuracy. Added Langfuse observability and migrated batch jobs to a Kafka event-driven architecture for zero-downtime inference."
    },
    {
      "role": "Associate Data Scientist",
      "company": "Bajaj Finserv Health",
      "location": "Pune, India",
      "dates": "Jul 2022 - Oct 2023",
      "summary": "Redesigned the production NER pipeline to boost fraud identification accuracy 27% while reducing false positives, and shipped a supervised model (logistic regression + feature engineering) that improved operational efficiency 22%."
    },
    {
      "role": "Data Engineer Intern",
      "company": "Bajaj Finserv Health",
      "location": "Pune, India",
      "dates": "Jan 2022 - Jun 2022",
      "summary": "Scaled ETL and analytics to 10M+ records across 200+ features by designing a distributed system (C++, Trino, Docker) with optimized SQL, analytical query processing, and disaster recovery."
    }
  ],
  "projects": [
    {
      "slug": "agentic-trend-orchestrator",
      "name": "Trend Pilot",
      "alternateName": "Agentic Trend Orchestrator",
      "tagline": "Agentic AI assistant for creator teams - catches trends, decides what to film, and assigns the work. Multi-agent system with GPT-4o VLM trend detection.",
      "date": "May 2026 - Jun 2026",
      "tech": ["FastAPI", "Next.js", "Docker", "Python", "TypeScript", "GPT-4o VLM", "Milvus", "RAG", "Multi-Agent", "CI/CD"],
      "links": {
        "demo": "https://trendpilot.shamik-basu.com/",
        "github": "https://github.com/ShamikOfficial/agentic-trend-orchestrator",
        "report": "assets/documents/Trend Pilot Deck.pdf"
      },
      "youtube": "",
      "video": "assets/video/Trend_Pilot_Intro.mp4"
    },
    {
      "slug": "ecomate-ai",
      "name": "EcoMate-AI",
      "tagline": "1st-place multimodal agentic platform turning receipts and activity into sustainability actions.",
      "date": "May 2025 - Jun 2025",
      "award": "1st Place - UCLA SAIRS Hackathon 2025",
      "tech": ["Gemini 2.5 Flash", "Python", "Streamlit", "Computer Vision", "NLP", "RAG"],
      "links": {
        "demo": "https://ecomateai.streamlit.app/",
        "github": "https://github.com/ShamikOfficial/EcoMate-AI",
        "devpost": "https://devpost.com/shamik1900"
      },
      "youtube": "tYTzNmGuq_E"
    },
    {
      "slug": "custom-cuda-library",
      "name": "Custom CUDA Library for CNN Pre-Processing",
      "tagline": "Open-source CUDA library for CNN preprocessing - shared-memory tiled kernels, Python .so bindings, 115x CPU speedup on Tesla T4.",
      "date": "Jan 2026 - Feb 2026",
      "tech": ["CUDA", "C/C++", "cuBLAS", "Python", "ctypes", "NumPy", "GPU Computing"],
      "links": {
        "github": "https://github.com/ShamikOfficial/CUDA-Accelerated-Python-Library",
        "report": "assets/documents/CUDA_Project_Report_Final.pdf"
      },
      "youtube": ""
    },
    {
      "slug": "smooth-operation",
      "name": "Smooth Operation",
      "tagline": "Turns Atlassian workflow chaos into clarity with real-time flow graphs and heatmaps.",
      "date": "2025",
      "tech": ["Python", "AI", "Data Visualization", "Atlassian APIs"],
      "links": {
        "devpost": "https://devpost.com/shamik1900"
      },
      "youtube": "P2oTHZWDBxU"
    }
  ],
  "education": [
    {
      "school": "University of Southern California",
      "degree": "Master of Science, Applied Data Science",
      "location": "Los Angeles, CA",
      "dates": "Jan 2025 - Dec 2026",
      "gpa": "3.7 / 4.0",
      "coursework": "Deep Learning, Machine Learning, Data Mining (MapReduce), Data Management"
    },
    {
      "school": "SRM Institute of Science & Technology",
      "degree": "Bachelor of Technology, Computer Science",
      "location": "Chennai, India",
      "dates": "May 2018 - Jun 2022",
      "gpa": "3.46 / 4.0",
      "coursework": "Algorithms, Artificial Intelligence, Computer Vision, Probability & Queueing Theory"
    }
  ],
  "skills": {
    "Languages & Databases": ["Python", "SQL", "R", "C/C++", "PostgreSQL", "MongoDB", "Spark", "Trino", "Apache Kafka"],
    "ML & Analytics": ["Machine Learning", "Statistics", "Econometrics", "Causal Inference", "A/B Testing", "Feature Engineering", "NLP", "NER", "Forecasting", "Tableau", "Power BI"],
    "Deep Learning & LLMs": ["PyTorch", "TensorFlow", "Hugging Face", "LoRA Fine-Tuning", "RAG", "LangChain", "Multi-Agent Orchestration"],
    "Cloud & DevOps": ["AWS", "Docker", "Kubernetes", "REST APIs", "CI/CD", "Azure Synapse", "Langfuse", "Git"]
  },
  "leadership": [
    {
      "role": "Vice President",
      "org": "GRIDS - Graduate Data Science Club, USC",
      "dates": "Sep 2025 - Present",
      "summary": "Lead analytics workshops, ideathons, and projects for a 250-member graduate data science community."
    },
    {
      "role": "Graduate Mentor",
      "org": "USC Viterbi School of Engineering",
      "dates": "Jan 2025 - Present",
      "summary": "Mentor graduate students on data science career paths, technical communication, and professional development."
    }
  ]
}