Ajay Bendale

About Ajay Bendale

Senior Data Engineer focused on building reliable, scalable data platforms that power analytics and machine learning in production.I have 4+ years of experience designing and operating batch and near–real-time data pipelines that process 10M+ records per day, with a strong focus on data quality, performance, and system reliability. My work sits at the intersection of data engineering and ML infrastructure, enabling teams to move from raw data to production-ready features and inference pipelines.I specialize in Python, SQL, PySpark, and Polars, and have hands-on experience building distributed data pipelines on Azure, containerized with Docker, and deployed across cloud and edge environments. I’ve worked extensively on incremental processing, schema evolution, data versioning, and idempotent pipeline design, ensuring reproducible and auditable datasets across training and inference workflows.I’ve led and contributed to systems that:Process high-frequency industrial data at scaleSupport near–real-time ML inferenceImprove pipeline performance by 15–20%Maintain reliability through validation, retries, and failure isolationEnable modular, plug-and-play data transformationsI also evaluate and prototype modern data platforms (e.g, Azure Fabric) to guide architectural decisions, balancing scalability, operability, and long-term maintainability.I enjoy working on problems that sit close to data platform architecture, distributed systems, and ML data infrastructure, and collaborating with product, ML, and platform teams to ship production-grade systems. Inventor on 5 business patents in automation and data engineering Recognized with 15+ innovation and performance awardsCore areas:Data Engineering · ETL/ELT · Distributed Systems · ML Data Pipelines · Python · PySpark · Polars · SQL · Azure · Docker · Data Modeling · Performance Optimization · Pipeline Reliability

Experience

1. System Engineer

   Tata Consultancy Services

   Nov 2024 — Present · Bengaluru, IN

   Own the design and operation of near–real-time data pipelines processing 10M+ records per day, supporting production ML inference workloads deployed on edge servers.Build and maintain high-performance preprocessing frameworks using Python, Polars, and parallel execution, reducing pipeline runtime by 15–20%.Design modular, plug-and-play data pipelines that allow new ML models and transformations to be onboarded without changes to core orchestration logic.Implement idempotent, fault-tolerant workflows with validation, retries, and failure isolation to ensure reliable execution in production environments.Containerize data services using Docker, enabling reproducible deployments across cloud and edge environments.Collaborate closely with ML engineers to deliver versioned, reproducible datasets used for model inference and evaluation.Lead evaluation of modern data platforms (including Azure Fabric) by assessing orchestration behavior, scalability, and integration trade-offs to guide long-term architecture decisions.Contribute to architectural discussions around data reliability, performance optimization, and system scalability.

Education

• MGM’s Jawaharlal Nehru Engineering College

  Bachelor of Technology - BTech, Mechanical Engineering

  2017

• Vasantrao Naik College, Aurangabad, Maharashtra.

  High School, Engineering Science

  2015 — 2017

• Sheth Lalji Narayanji Sarvajanik Vidyalaya, Jalgaon, Maharashtra

  Secondary chool

  2005 — 2015