Headline

Senior ​Data Engineer | Python, PySpark, Azure Databricks | LLM Agents, LangChain, Model Serving & Databricks Apps | DP-203 & AI-102 Certified | Bridging Data & Applied AI

About

I turn legacy, high-maintenance data systems into cloud-native platforms that teams can trust. At KPMG's Global Delivery Centre, I rebuilt three legacy Informatica pipelines into Python and Databricks frameworks, cutting maintenance overhead by around 60%, query latency by around 45%, and automating roughly 70% of manual workflows across financial and audit data. I am a Data Engineer specialising in cloud-native, scalable data platforms built on Azure Databricks, Delta Lake, and PySpark. Over the past three years, I have engineered, optimized, and automated enterprise ETL pipelines powering analytics and audit operations for 200+ users across high-governance financial domains. Beyond traditional pipeline engineering, I focus on bridging robust backend data infrastructure with production-grade applied AI and solutions engineering. My recent work expands directly into Forward Deployed and AI Engineering—building agentic workflows and AI products end-to-end. This includes developing custom LLM agents using LangChain (leveraging Opus 4.8), orchestrating model endpoints via Databricks Model Serving, and delivering interactive user-facing tools using Databricks Apps. Certified as both an Azure Data Engineer Associate (DP-203) and Azure AI Engineer Associate (AI-102), I thrive on taking complex architectures and turning them into scalable, automated, and high-impact business solutions. Core toolkit: Azure Databricks, Delta Lake, PySpark, Python, Git, Workflow Automation, SQL, Power BI, Power Automate, CI/CD, LangChain, LLM Agents, Databricks Model Serving, Databricks Apps, Agentic Workflows. I am currently open to Data Engineering, FDE and Applied AI roles, including opportunities open to US relocation as well as remote positions. If your team is hiring or you would just like to connect, reach me at harshsachde0@gmail.com.

Location

Mumbai Metropolitan Region

AI stack

claude code

claude

gemini

github copilot

openai

Microsoft Copilot

Work experience

Senior Data Engineer at KPMG, Oct 2026 - Present · 1 mo, Bengaluru, Karnataka, India · Hybrid, - Replaced three legacy Informatica IICS pipelines with scalable Python and Databricks ETL frameworks across multiple source tables, cutting maintenance overhead by around 60% and simplifying long-term support. - Spearheaded the end-to-end development of production AI applications on Databricks—building autonomous LLM agents using LangChain (leveraging Opus 4.8), hosting endpoints via Databricks Model Serving, and delivering interactive internal tools using Databricks Apps. - Built PySpark notebooks and Databricks Jobs to ingest and transform SAP ECC6 General Ledger and trial-balance data into Delta Lake, using incremental ingestion and optimised partitioning to reduce query latency by around 45%. - Automated SharePoint approval and CRUD workflows with Power Automate, removing manual steps and automating roughly 70% of a recurring audit-support process. - Curated and validated audit-ready datasets for MindBridge anomaly analysis, giving U.S. Audit stakeholders trusted, well-modelled data. Led performance-tuning and data-observability work (logging, lineage, alerting) across pipelines and mentored junior engineers. Recognised twice as Associate of the Quarter.

Data Engineer at KPMG, Jun 2023 - Oct 2026 · 3 yrs 5 mos, Bengaluru, Karnataka, India · Hybrid, - Replaced three legacy Informatica IICS pipelines with scalable Python and Databricks ETL frameworks across multiple source tables, cutting maintenance overhead by around 60% and simplifying long-term support. - Spearheaded the end-to-end development of production AI applications on Databricks—building autonomous LLM agents using LangChain (leveraging Opus 4.8), hosting endpoints via Databricks Model Serving, and delivering interactive internal tools using Databricks Apps. - Built PySpark notebooks and Databricks Jobs to ingest and transform SAP ECC6 General Ledger and trial-balance data into Delta Lake, using incremental ingestion and optimised partitioning to reduce query latency by around 45%. - Automated SharePoint approval and CRUD workflows with Power Automate, removing manual steps and automating roughly 70% of a recurring audit-support process. - Curated and validated audit-ready datasets for MindBridge anomaly analysis, giving U.S. Audit stakeholders trusted, well-modelled data. Led performance-tuning and data-observability work (logging, lineage, alerting) across pipelines and mentored junior engineers. Recognised twice as Associate of the Quarter.

Consultant at TalaKunchi Networks Pvt Ltd, May 2022 - Sep 2022 · 5 mos, Researched cybersecurity compliance and audit frameworks (RBI, SEBI, ISO/IEC 27000) and documented control gaps, strengthening client audit readiness.

Application Developer at Finbits India Pvt. Ltd., Feb 2021 - Jun 2021 · 5 mos, Built a back-end portal to manage raw JSON for API callbacks and integrated it with a Flutter app, cutting manual JSON handling and speeding up testing cycles.

Education

Integrated MBA Tech (B.Tech in Information Technology + MBA in Technology Management), Information Technology, SVKM's Narsee Monjee Institute of Management Studies (NMIMS)

Public links

github: https://github.com/harshsachde