Case studies across product engineering and applied AI

Selected Work

I work best on products that need both strong engineering fundamentals and pragmatic AI. That usually means shipping under production constraints, not just building prototypes.

2024 - Present

Software Development Engineer 2

IBM Software Labs · Bengaluru, India

I design and ship reliability-heavy AI capabilities inside browser automation and testing products. The work combines embeddings, vision-language models, service decomposition, and a lot of operational discipline.

  • Designed AI-powered auto-healing with a three-tier recovery ladder: CSS selectors, text embeddings, then IBM Granite 3.3 VLM.
  • Built logic that autonomously corrects locator failures across 5K+ test cases with 100% Firefox accuracy and 83% Chrome accuracy in internal evaluation.
  • Helped re-architect the SAT runtime from a monolith into four microservices handling 4K-5K requests per minute with sub-second latency.
  • Led a Java 8 to 17 migration and CI/CD improvements that improved build speed by 30% while maintaining 99% uptime.

2023 - 2024

Software Engineer 2

Software AG (now IBM) · Bengaluru, India

This phase pushed me deeper into AI product work: semantic retrieval, internal copilots, and prediction systems grounded in practical product needs rather than demos.

  • Built a semantic search engine using NLP, knowledge graphs, and FAISS. The project won the TechInterrupt Hackathon: first in India and fourth internationally.
  • Developed an AI chatbot with LangChain and Flask that reduced internal support tickets by 70%.
  • Created a failure prediction system using PyTorch and time-series analysis that supported 99.99% system availability.

2022 - 2023

Software Engineer

Software AG · Bengaluru, India

I worked on enterprise integration platform capabilities across Java, Spring Boot, and REST APIs, building the foundation that still shapes how I reason about production systems.

  • Delivered webMethods platform features for enterprise integration use cases using Java and Spring Boot.
  • Built with a strong emphasis on compatibility, API contracts, and release quality, and was recognized as a 2023 Star Performer.

How I operate

Start with the cheapest reliable fallback

I design systems to attempt deterministic recovery first, then graduate into ML and model-based fallbacks only when they are justified.

Treat evaluation as part of the product

When the system contains AI, the measurement loop is not optional. I care about observable accuracy, drift, error budgets, and failure analysis.

Bias toward boring operations

I prefer architectures that are easier to debug, deploy, and recover under pressure over clever stacks that only look good in diagrams.

Ship ownership, not isolated demos

The work that matters is the work that survives adoption. I optimize for maintainability, team adoption, and operational credibility.

Capability map

AI and ML

PyTorchTransformersLLM pretrainingText embeddingsKnowledge graphsLangChainComputer visionFAISS

Backend and platform

JavaPythonGoSpring BootFlaskREST APIsMicroservicesKafkagRPC

Operations

DockerJenkinsNGINXCI/CDLinuxReliability engineeringState managementRelease ownership

Interested in working together?

Let's discuss role fit

I am most interested in roles where AI systems, backend engineering, and reliability work intersect.

Get in touch →