👋 Hello, I'm Punith_A

Data Engineer | Aspiring IT Project Manager | AI Enthusiast

Data Engineer with 4+ years of experience designing and delivering enterprise-scale data platforms across financial services, insurance, retail, and telecom sectors. I specialize in building robust, cloud-native data architectures using modern lakehouse patterns translating raw data into reliable, high-performance analytical systems.

I've driven significant cost and performance wins including Snowflake compute cost reductions, query latency improvements, and large-scale cloud migration projects. My engineering philosophy: build for reliability, optimize for cost, and scale for the future.

Enterprise banking data platform work across reporting, analytics, and pipeline engineering.

Retail data platform engineering; supply chain and POS analytics pipelines.

AI/ML data infrastructure; churn prediction feature pipelines and model data workflows.

graph TB
    A[Raw Sources:<br/>APIs · Kafka · Databases · Files] --> B[Ingestion Layer<br/>Spark · Glue · ADF · Custom ETL]
    B --> C[Cloud Data Lake<br/>S3 · ADLS · GCS · Delta Lake]
    C --> D[Transformation<br/>PySpark · Scala · dbt · SQL]
    D --> E[Serving Layer<br/>Snowflake · Redshift · BigQuery]
    E --> F[Consumption<br/>Power BI · Tableau · ML Models · APIs]

    style A fill:#0f172a,stroke:#3b82f6,stroke-width:2px,color:#fff
    style B fill:#1e40af,stroke:#60a5fa,stroke-width:2px,color:#fff
    style C fill:#1d4ed8,stroke:#93c5fd,stroke-width:2px,color:#fff
    style D fill:#2563eb,stroke:#bfdbfe,stroke-width:2px,color:#fff
    style E fill:#3b82f6,stroke:#dbeafe,stroke-width:2px,color:#fff
    style F fill:#059669,stroke:#6ee7b7,stroke-width:2px,color:#fff

timeline
    section Early Career
        Foundation : Junior to Mid-level Data Engineering
                   : ETL · SQL · Warehousing basics
    section Growth
        Voziq AI     : Data Engineer
                     : Feature engineering for churn models
The Hanover Insurance: Retail analytics platform
                     : POS and supply chain pipelines
    section Senior
        SMBC       : Enterprise banking platform
                   : Snowflake optimization · Cloud migration
                   : 2024–Present

End-to-end ML pipeline built on the UCI CKD dataset. Focused on clinical interpretability using SHAP values, odds ratios, and decision tree visualizations — designed for healthcare stakeholder communication, not just model accuracy.

Stack: Python · Scikit-learn · SHAP · Pandas · Matplotlib

Terminal-aesthetic developer tool for GitHub profile analysis and optimization. Built with a polished UI targeting engineers who want to sharpen their personal brand and portfolio positioning.

Stack: Python · GitHub API · Rich CLI

Multiple enterprise cloud migration projects across AWS, GCP, and Azure — moving legacy on-premise data systems to modern lakehouse architectures with measurable latency and cost improvements.

Stack: Spark · Databricks · Snowflake · Airflow · Delta Lake · dbt

"Great data engineering is invisible — pipelines run silently, warehouses respond instantly, and business teams make confident decisions without wondering where the numbers come from. That's the standard I build to."

Core principles I ship by:

⚡ Performance-first — Profile before optimizing; instrument everything
🔁 Idempotency by default — Every pipeline should be safely re-runnable
💸 Cost as a metric — Cloud spend is an engineering responsibility, not just finance's
📐 Schema as contract — Data contracts prevent downstream chaos
🧪 Test your data — Data quality checks are not optional in production