Lokesh Nanda

Print or save as PDF

Lokesh Nanda

Principal Architect — Cloud, Data & AI · Bengaluru, India

Engineer and architect with 12+ years of experience designing data platforms, cloud solutions and AI products. I bridge the gap between technical teams and business stakeholders — translating complex problems into systems that ship. I’ve built and led high-performing engineering teams, delivered platforms serving millions of customers, and driven cloud strategy across GCP, AWS and Azure. On the side, I build apps and demos you can try on this site.

What I work with

  • Data engineering: Apache Spark, Apache Beam, Kafka, Airflow, BigQuery, lakehouse architectures (Microsoft Fabric), Hadoop ecosystem
  • Cloud: GCP, AWS and Azure — architecture, migration and cost optimisation
  • AI products: LLM chatbots (Azure OpenAI, Azure AI Search), multi-agent systems (AutoGen), AI-assisted tooling
  • Engineering platforms: Python, SQL, FastAPI, microservices, Docker, Kubernetes, Terraform, GitLab/Azure DevOps CI/CD
  • Leadership: hiring and building teams, architecture strategy, client engagement

Experience

SG Analytics — Principal Architect

May 2026 – present · previously Senior Solutions Architect, May 2024 – April 2026

  • Created the cloud strategy for SG Analytics’ in-house projects and reduced cloud costs across Azure, GCP and AWS by 40%.
  • Hired and lead a core tech team of 12 data engineers and data scientists for the Data & Analytics line of business.
  • Built an AI chatbot integrated with Microsoft Teams (Microsoft Bot Framework, Azure AI Search, Azure OpenAI) and scaled it to 800 users in production.
  • Architected and built a lakehouse on Microsoft Fabric; developed REST APIs with FastAPI, Azure APIM and Function Apps.
  • Building multi-agent systems with the AutoGen framework, deployed on Azure.
  • Introduced a DevOps approach to client delivery and set up the DevOps team, contributing to improved NPS.

Falabella India — Senior Architect

July 2023 – April 2024

  • Led cloud cost reduction for the Falabella group: built the team, set the strategy, and ran GCP best-practice sessions across the group — cutting overall cloud spend by 30%.
  • Replicated the cost-optimisation model across two other business arms (Sodimac and Tottus), with dashboards and alerts for ongoing cost control.
  • Redesigned ingestion and transformation of web-navigation data from Adobe Omniture, fixing long-standing data-quality issues.

Falabella India — Architect

October 2022 – June 2023

  • Implemented data lineage and observability for ingestion pipelines using Cloud Dataplex and Analytics Hub.
  • Data architect for Falabella.com search — treating search data as a service; improved the Learning-to-Rank pipeline with Spark for better reliability and scale.
  • Brought CI/CD (GitLab) and Terraform infrastructure-as-code to all data products — every pipeline gained one-click deployment.

Falabella India — Lead Data Engineer

April 2021 – September 2022

  • Built the Customer Engagement Platform (CEP): a one-stop campaign-management platform unifying marketing across Falabella’s businesses.
  • Designed and developed the platform’s API layer with FastAPI, integrating Salesforce, Airship, Masiv and Digevo.
  • Hired the core team, and worked with data scientists on analytical products — campaign measurement, target-audience generation and contact-policy optimisation.

Falabella India — Senior Data Engineer

December 2019 – March 2021

  • Core engineer on Falabella’s Customer Data Platform: generated a unified customer ID for 67 million customers across 7 business units using Apache Beam.
  • Managed 60 TB of customer data across 120+ entities and 500+ pipelines (BigQuery, Beam, Spark, Python, Airflow), with Looker dashboards on top.
  • Platform served 7 business types and 760+ users; ran workshops with business teams to drive adoption.

Lowe’s Companies Inc. — Senior Analyst

January 2018 – December 2019

  • Built a Python framework to gather and analyse contact-centre customer-feedback data from InMoment and Medallia, giving a 360° view of customer satisfaction.
  • Cut a 3-hour audio-transcript processing job to 10 minutes using multithreading; automated deployments with Jenkins.
  • Designed a generic data-ingestion framework used across ingestion teams, with incremental loads and robust logging.

Brillio Technologies — Senior Data Engineer

April 2016 – January 2018

  • Key designer of a security-monitoring and log-management platform for Bristol Myers Squibb on AWS: Kinesis streaming to S3 via Lambda, Athena partitioning, DynamoDB-backed logging and email alerting.
  • Extensive automation with the AWS boto3 API; explored Glue to replace the existing pipeline framework.

Brillio Technologies — Data Engineer

April 2014 – April 2016

  • Set up a Hortonworks Hadoop cluster on Azure; parsed millions of XML files with Spark into Hive for analytics.
  • Built core components of the Brillio Data Platform: social-media connectors, real-time Kafka–Spark analytics demos, a Google Analytics crawler, and one-click cluster automation; supported technical pre-sales with client teams.

Selected side projects

  • Family Health Tracker — health records app for families
  • OSS Radar AI — AI-assisted open-source discovery
  • Rep Logs — friction-free workout logging
  • Book Notes AI — book notes with AI summaries
  • Interactive demos — dashboards built with realistic sample data across banking, retail, public safety, elections, fintech and IT operations: full collection
  • This website — publishes itself automatically from markdown and git

Certification

  • Google Cloud Certified Professional Cloud Architect (2023)

Education

  • Bachelor of Engineering (EEE) — BMS Institute of Technology, VTU, 2013

Writing & community

Languages

English · Hindi · Odia

Contact