AI SME / Team Lead — AI Engineer with 5 years of experience delivering AI and GenAI solutions for U.S. healthcare and Fortune 100 clients. Expertise spans AI solution architecture, client engagement, AI agents, RAG, OCR, document intelligence, LLM applications, and workflow automation using Azure AI Foundry, LangChain, LangGraph, Python, Django, FastAPI, and Azure services. Proven expertise in leading teams, translating business requirements into scalable AI solutions, and leveraging a strong data engineering background across Azure, Databricks, Spark, Hadoop, and large-scale data platforms.
LinkedIn GitHub Resume
Leading AI and GenAI initiatives directly with TCS within the TCS Healthcare Innovation Lab, partnering with multiple U.S. healthcare clients to understand business requirements, design solution architectures, and deliver scalable AI solutions across diverse healthcare use cases. Engaging with client stakeholders to gather requirements, translate business problems into technical solutions, and lead end-to-end delivery in collaboration with the engineering team. Driving solution development from discovery and architecture design through prototyping, implementation, and delivery, while guiding the team on AI technologies, solution design, and execution.
Architecting and delivering AI applications including OCR, medical document extraction, intelligent text processing, information extraction, and workflow automation using Azure AI Foundry, LangChain, and Azure AI services. Partnering directly with multiple U.S. healthcare clients to gather requirements, translate business problems into technical solutions, and lead end-to-end delivery with the engineering team.
Developing scalable full-stack AI applications using Azure AI Foundry, LangChain, Azure Functions, Django, React, and Azure cloud services, integrating AI capabilities into enterprise healthcare workflows. Leading technical initiatives across discovery, design, implementation, and delivery within the TCS Healthcare Innovation Lab.
Delivered a portfolio of AI-powered healthcare solutions including Auto RFP, MR Extractor, handwritten prescription OCR, and discharge bill processing, along with numerous client-specific AI use cases. Leveraging a strong data engineering background across Azure, Databricks, Spark, Hadoop, and large-scale data platforms to support implementation.
Delivered 20TB+ of healthcare data into Azure Data Lake Storage (ADLS) by designing and optimizing enterprise-scale data pipelines using Azure Data Factory (ADF) and Databricks Delta Lake, enabling faster analytics and reporting. Increased data processing efficiency by 60% by automating ETL/ELT workflows across ADF, ADLS, Databricks, and PySpark, ensuring near real-time data availability. Improved data accuracy by 100% through implementing robust validation and quality frameworks, delivering highly reliable and actionable business insights.
Enabled seamless enterprise-wide analytics by integrating on-premise, third-party, and Hadoop data sources (Oracle, Taleo, JSON, Avro, CSV, Parquet) into ADLS via ADF connectors. Reduced Spark job runtime by 20% for 2M+ daily records by implementing Spark performance tuning techniques (caching, partitioning, broadcast joins) across Hadoop environments. Optimized on-premise Hadoop clusters (HDFS, Hive, Spark) by fine-tuning resource allocation and data partitioning strategies, improving large-scale batch processing performance by 35% and ensuring smooth integration with Azure cloud migration pipelines.
Accelerated large-scale text processing by 90% (2 hours → 15 minutes) by developing and optimizing MapReduce programs in the Hadoop ecosystem, significantly improving ETL throughput.
Enhanced operational efficiency by 50% by automating data workflows using Python, HiveQL, and Shell scripting, reducing manual intervention and improving reliability.
Improved analytical data trustworthiness by 30% through data quality, consistency, and reconciliation checks using Hive, Impala, and Spark.
Streamlined ETL orchestration and reduced manual efforts by 40% by automating complex data pipelines with Python, HiveQL, and Shell, achieving a 60% improvement in throughput.
Supported cloud migration initiatives by managing data ingestion and batch processing in Hadoop Distributed File System (HDFS), laying the foundation for Azure-based architectures. Delivered enterprise-wide reporting solutions by collaborating with business teams and processing large-scale datasets across distributed ETL systems. Automated complex data pipelines with Python, HiveQL, and Shell scripting, achieving a 60% improvement in throughput. Streamlined ETL orchestration and reduced manual efforts by 40%, significantly improving reliability across distributed systems.
Himanshu has developed and published two open-source Python libraries that address real-world challenges in NLP retrieval and big data job metadata analysis. Both projects are fully documented, available on PyPI, and designed to be privacy-friendly, zero-dependency alternatives to cloud-based or complex enterprise solutions.
Developed and published an open-source Python library, liuembeddings, which combines text embeddings and vector storage for efficient local NLP retrieval tasks. Built on Hugging Face Transformers and ChromaDB, it offers a free, fully local, and privacy-friendly alternative to cloud-based embedding services — ideal for AI projects requiring fast and cost-efficient embedding, storage, and retrieval workflows.
Developed and published an open-source, zero-dependency Python engine, DbMeta, designed for parsing and analyzing batch job metadata from Hadoop, Spark, and Databricks by using folder-based job-run structures. Automatically detects metadata files, extracts structured metadata, and converts them into queryable tables for efficient monitoring and analytics.
Himanshu's technical expertise spans the full AI and data engineering stack — from designing GenAI solutions with Azure AI Foundry, LangChain, and LangGraph, to building backend APIs with Python, Django, and FastAPI, and architecting enterprise-scale data pipelines on Azure, Databricks, and Hadoop. His skills cover AI & GenAI, backend & APIs, cloud & Azure services, vector & data stores, data architecture, and data engineering.
Azure AI Foundry, LangChain, LangGraph, Prompt Engineering, Open-Source Models
Python, Django, Django REST Framework, FastAPI
Azure Functions, Azure Blob Storage, Azure Queue Storage, ADLS, Azure DevOps
MongoDB, SQL, PostgreSQL, MySQL, Oracle
ETL/ELT, Data Modelling, Data Warehousing, Data Lakehouse, OLAP/OLTP
Databricks, Apache Spark (PySpark, RDD, Streaming), Hadoop, Hive, Impala, MapReduce, ADF, Azure Synapse
Himanshu holds certifications spanning data engineering, cloud platforms, and generative AI, reflecting a continuous commitment to professional development across the technologies he applies in production systems for enterprise clients.
A collection of open-source projects showcasing expertise in PySpark, SQL translation, Databricks analytics, ETL pipelines, and AI-powered chatbots. Each project demonstrates hands-on proficiency with real-world data engineering and AI challenges, from translating LeetCode and DataLemur SQL problems into PySpark and RDD implementations, to building ecommerce analytics dashboards on Databricks, implementing end-to-end ETL pipelines, and developing AI chatbots and analysis bots. These projects reflect a consistent commitment to sharing practical solutions with the broader data engineering and AI community.
Computer Science & Engineering
NIET — Greater Noida, UP
Aug 2017 – Sep 2021
ABC Public School — Gorakhpur, UP
Apr 2015 – Mar 2017
Himanshu's academic foundation in Computer Science & Engineering from NIET provided the theoretical and practical groundwork for his career in data engineering and AI. His educational background, combined with continuous professional certifications in Databricks, Azure, and Generative AI, reflects a sustained commitment to staying current with evolving technologies in the data and AI space.
With 5 years of experience delivering AI and GenAI solutions for U.S. healthcare and Fortune 100 clients, Himanshu brings proven expertise in AI solution architecture, client engagement, AI agents, RAG, OCR, document intelligence, LLM applications, and workflow automation. His background spans leading teams at TCS Healthcare Innovation Lab, architecting enterprise-scale data pipelines for Kaiser Permanente, and optimizing large-scale ETL systems for Walgreens — all backed by a strong data engineering foundation across Azure, Databricks, Spark, and Hadoop.
Whether you're looking to explore AI solution architecture, GenAI implementation, data engineering, or team leadership for enterprise healthcare and Fortune 100 projects, Himanshu is open to connecting and discussing opportunities. Reach out via LinkedIn, GitHub, or email to start a conversation about how his experience with Azure AI Foundry, LangChain, LangGraph, Python, Django, FastAPI, and Azure services can support your organization's goals.