Abhinav Sreesan

Overview

Senior Data Engineer based in Mumbai with hands-on experience designing, building, and operating enterprise multi-cloud data platforms and analytics infrastructure.

See experience and projects for implementation details and delivery outcomes.

Experience

Projects

Project highlights from client platforms and internal accelerators aligned to production impact.

Transaction Intelligence & Consumer Analytics Feed

Engineered a high-scale data intelligence engine that leverages Large Language Models to normalize fragmented card transactions, providing institutional investors with daily insights into company health and consumer spend patterns.

What I delivered

• Designed an LLM-driven normalization engine to clean and resolve unstructured transaction descriptions into a standardized format.
• Built a sophisticated vendor-tagging system that maps cleaned transactions to an enterprise “Vendor Universe” using semantic entity resolution.
• Developed a panelization framework to transform raw transactional data into statistically significant cohorts for trend analysis.
• Delivered a high-velocity daily feed providing granular metrics on company growth, spend velocity, and market-share shifts.
• Partnered with research analysts to define KPIs that quantify company health and consumer loyalty from raw financial signals.

Platform capabilities

• Scaled processing on Databricks to handle millions of daily records using distributed PySpark workflows.
• Orchestrated LLM integration for automated text classification and entity mapping at scale, significantly reducing manual tagging.
• Implemented panel-based normalization logic to account for data bias and ensure accurate representative spend modeling.
• Built automated data quality gates and drift monitoring to ensure the integrity and reliability of the daily financial feed.

Command and Control Data Platform

Delivered a centralized “Command and Control” platform that provided stakeholders with real-time visibility across manufacturing, sales, and logistics workflows.

What I delivered

• Architected an Azure Synapse environment with complex pipelines orchestrating end-to-end ingestion and transformation journeys.
• Strategized with business stakeholders to map disparate manufacturing and logistics workflows into a cohesive, unified data model.
• Engineered high-performance integration patterns for SAP CDC and API sources using PySpark on Databricks.
• Scaled the pilot to support 40+ critical KPIs, providing granular drilldowns for over 30 manufacturing plants.
• Led a cross-functional squad of four engineers to successfully deliver the production pilot within the first month.

Platform capabilities

• Governance Framework: Built a metadata-driven engine using Cosmos DB and serverless Azure Functions for automated scheduling and configuration.
• Automated Observability: Implemented proactive data quality monitoring with benchmark deviation alerts and automated escalation routing.
• Secure DevOps: Established robust CI/CD workflows for the entire stack and designed a custom row-level security (RLS) framework for fine-grained access.

Metadata and Data Quality Framework

Built a reusable framework-first operating model that transformed raw infrastructure into an observable and governed data platform.

Highlights

• Metadata Engine: Developed a centralized source configuration and scheduling hub using Cosmos DB with a serverless Function API.
• Quality Observability: Engineered automated quality benchmark monitoring in Synapse, featuring deviation alerting and intelligent escalation logic.
• Unified Security: Designed a custom row-level security architecture to enforce strict data privacy across diverse business units.
• Centralized Logging: Consolidated pipeline logging and audit trails to improve troubleshooting speed and platform transparency.

Delivery & Automation

• Automated Lifecycle: Configured end-to-end CI/CD pipelines for Synapse, Cosmos DB, and Azure Functions to ensure reliable, repeatable deployments.

In-House Data Platform Accelerator

Designed a reusable deployment accelerator that allows teams to provision standardized data platforms in minutes rather than weeks.

What it automated

• Infrastructure-as-Code: Provisioned integrated Azure stacks (Synapse/ADF, Databricks, and ADLS) using reusable Terraform templates.
• Environment Orchestration: Enabled seamless multi-environment deployments via GitHub Actions and Azure DevOps pipelines.
• Standardized Architecture: Baked in best-practice security and networking configurations by default for every new deployment.

Impact and innovations

• Operational Efficiency: Slashed configuration and deployment time by 60%, significantly accelerating project kickoff phases.
• Architecture Exploration: Evaluated the use of Delta tables for metadata storage to further decouple compute from governance databases.
• Future-Proofing: Built a Delta Catalog POC to explore advanced data sharing and enhanced RBAC controls within the Databricks ecosystem.

Skills

Tools and methods used across ingestion, transformation, governance, and delivery automation.

Technologies

Techniques

Certifications

• Databricks Certified Data Engineer Professional (Feb 2024)
• DP-700 Microsoft Certified: Fabric Data Engineer Associate (Jul 2025)
• DP-201 Designing an Azure Data Solution (Jan 2021)
• DP-200 Implementing an Azure Data Solution (Feb 2020)

Academic Achievement

Technical paper publication (2019): “Machine Learning Based Autonomous Road Maintenance System Using Cold Lay Asphalt” in the international journal Helix, Vol. 8(5): 4002-4006.

Education

University of Mumbai, Mumbai - Bachelor of Engineering in Electronics and Telecommunication (Jun 2018).

Contact

Contact me via email or LinkedIn. Happy to connect and discuss data platform engineering, architecture, and delivery.

• abhinavsreesancp@outlook.com
• LinkedIn

Elements

i = 0;

while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';