Shantanu Singh

Senior Research Associate with 3+ years of experience designing and deploying end-to-end AI/ML and Generative AI solutions across domains including healthcare, compliance, fintech, media, and digital services. Strong expertise in LLMs and RAG-based architectures, multi-Agent, including model fine-tuning and optimization using GPT, LLaMA, Mistral, Claude, and Amazon Titan to drive automation and intelligent decision-making. Hands-on experience across both AWS and Azure cloud ecosystems, building enterprise-grade AI platforms using services such as Amazon Bedrock, Amazon SageMaker, Azure OpenAI, Azure AI factory and Azure AI Search.

Proven ability to architect for scalable ML pipelines, implement LLM observability, and deploy real-time production-ready AI systems.

Senior Research Associate – AI/ML - CloudThat Technology Pvt. Ltd.

(2025-10)

• Led end-to-end delivery of high-impact Generative AI solutions across fintech, media, and SaaS domains.
• Designed and implemented Retrieval-Augmented Generation (RAG) knowledge systems using vector databases for accurate, context-aware information retrieval.
• Built and deployed multi-agent AI chatbot solutions to orchestrate complex workflows and decision-making processes.
• Architected enterprise-grade ML and LLM pipelines optimized for scalability, performance, and cost efficiency.
• Implemented LLM observability frameworks to monitor response quality, latency, hallucination risks, and operational costs.
• Mentored engineering teams on advanced prompt engineering techniques, GenAI architecture patterns, and AI solution design best practices.

Research Associate – AI/ML - CloudThat Technology Pvt. Ltd.

(2023-09 - 2025-10)

• Led the design and delivery of enterprise GenAI solutions, including a scalable multi-agent conversational AI platform, AI-powered cost sheet automation, and LLM-based chatbots for fintech, FMCG, and HR use cases.
• Built RAG-based knowledge systems using vector databases and metadata-driven ingestion pipelines to enable secure, multilingual, and context-aware responses.
• Optimized LLM inference and serverless pipelines (Amazon Bedrock, AWS Lambda) to reduce response latency and overall infrastructure costs.
• Integrated semantic search and recommendation logic across pharma and enterprise platforms to improve information discovery and personalization.

Research Intern - CloudThat Technology Pvt. Ltd.

(2023-01 - 2023-08)

• Deployed scalable AWS-based applications using S3, CloudFront, Glue, Lambda, and IAM.
• Conducted ETL operations and exploratory data analysis on diverse datasets.
• Configured VPCs, Route53 domains, and QuickSight dashboards.
• Built search indexes using Amazon Kendra and deployed ML inference endpoints.

B.Tech - Data Engineering - JIET institute of Design and Technology (2023)