

Technology Architect with 15 years of experience designing and scaling enterprise-grade AI/ML and Generative AI platforms. Deep expertise in GenAI system design, including LLM-based application architecture, prompt engineering, model fine-tuning, and Retrieval-Augmented Generation (RAG) solutions using Python, LangChain, and Hugging Face. Hands-on experience in developing ai agents for banking and financial use cases. Utilised airia.ai low-code orchestration and AI agent development platform effectively. Supported team efforts in creating innovative solutions for taxation use-cases based on observability, security, and cost optimization. Proven ability to translate complex business problems into robust AI solutions aligned with enterprise architecture and governance.
Completed a Post Graduate Diploma in Artificial Intelligence and Machine Learning from IIIT Bangalore, strengthening foundations in machine learning, data science, and applied AI through hands-on labs and industry-oriented projects.
Brings strong cross-platform engineering leadership from architecting enterprise-native iOS ecosystems across telecom, retail, aviation, mining, and insurance domains, enabling seamless AI integration into user-facing applications. Experienced in defining technical roadmaps, engineering standards and multi-team deliveries across AI and product engineering organizations.
Spearheading the architectural design and deployment of Generative AI Agentic Workflows tailored for high-stakes environments in Banking, Finance, and Taxation. Bridging the gap between cutting-edge LLM development and Responsible AI to deliver secure, ethical, and scalable enterprise solutions.
Key contributions:
Career Progression: System Engineer → IT Analyst (ITA) → Assistant Consultant (AST) → Associate Consultant (ASOC)
Contribution:
AI Solution Architect: Enterprise Atlassian Intelligence
Situation: US based retail brand faced data fragmentation across Jira and Confluence, hindering decision-making for a large data technical ecosystem.
Task: I architected a stateful AI assistant utilizing a self-hosted LLM and Vector DB design to automate reasoning over enterprise data silos.
Action: I deployed Mistral Large 3 via vLLM and Milvus within a private VPC, implementing prefix caching and reasoning items to deliver context-aware "opinions" on project blockers.
Result: Successfully scaled the solution to production, achieving a 40% reduction in ticket resolution time and significant operational cost savings.
AI/ML Solution for Global Telecom B2B Platform
Situation: A global B2B telecom platform required advanced analytics to mitigate churn and optimize customer usage patterns.
Task: I led the development of predictive ML models and structured LLM workflows to automate retention strategies.
Action: I built Python models for behavior prediction, engineered Chain-of-Thought prompts, and containerized the stack via Docker on Azure ML.
Result: Successfully deployed a scalable, multi-environment solution that improved prediction accuracy and streamlined data pipelines.
Retail Analytics B2B Platform
Situation: A B2B retail platform faced significant margin erosion and frequent stockouts due to manual inventory adjustments and static pricing across a 1TB product catalog.
Task: I led the evolution from legacy analytics to an autonomous Agentic AI framework to optimize demand forecasting and supply chain execution with minimal human intervention.
Action: I built Python-based forecasting models and architected a multi-agent system using LangGraph for stateful, cyclic orchestration. I designed "Planner" and "Executor" agents leveraging Chain-of-Thought reasoning and standardized function calling to trigger autonomous stock reorders and dynamic pricing updates. I integrated a RAG-powered BI agent using vLLM for high-throughput, context-aware Q&A, ensuring data sovereignty via a private VPC deployment.
Result: Successfully deployed the framework, resulting in 30% faster inventory turnover, a 20% reduction in stockouts, and a 12% increase in gross margins across all pilot categories.
Other iOS & Cross-Platform Projects
Contribution:
Contribution:
Situation: Existing crossword game needed expansion to macOS and richer interaction on iOS to boost engagement.
Task: Extend the app to desktop while enhancing mobile user experience and performance.
Action: Developed a macOS desktop version with platform-specific features, enhanced the iPhone app with gesture-driven interactions (drag-and-drop) and animations, integrated Game Center, and optimized code through performance analysis.
Result: Increased user engagement, improved gameplay responsiveness, and expanded the product’s reach across Apple platforms.