Jacquiline George

Generative AI Engineer with 3 years 10 months of experience building enterprise AI applications, RAG systems, and multi-agent solutions. Experienced across the AI application lifecycle from RLHF-style LLM data annotation and prompt engineering through backend development, agent orchestration, enterprise integrations, evaluation, and production deployment. Hands-on with Python, LangChain, LangGraph, FastAPI, AWS ECS, Amazon ECR, Amazon Bedrock AgentCore, Azure OpenAI, Azure Data Lake Storage, Snowflake, and Claude Code.

Currently improving a client-adopted AI application development harness supporting research, planning, building, and deployment.

AI & Data Science Engineer (Generative AI Engineer) at Deloitte Consulting — Offices of the USI (2022-12 – Present)

• Build and enhance enterprise Generative AI and Agentic AI applications, progressing from RLHF-style LLM data annotation and prompt engineering to RAG systems, multi-agent architectures, enterprise integrations, and production deployment.
• Currently enhance a Claude Code-based AI application development harness adopted by clients, supporting the end-to-end AI application lifecycle across research, planning, development, and deployment.
• Own the harness deployment phase, enabling applications to configure required deployment resources and deploy to Amazon ECR, AWS ECS, and Amazon Bedrock AgentCore using dedicated engineering skills; AWS CodeBuild handles container image builds.
• Conducted a firmwide deep-dive session demonstrating how teams can use the harness to configure and deploy AI applications to AWS ECS and Amazon Bedrock AgentCore.
• Designed and implemented the backend for multi-agent applications with multiple specialized agents and PostgreSQL-based agent-state management; integrated Snowflake agents with the platform so agents across both environments can communicate and participate in coordinated workflows.
• Designed and implemented the backend for a financial-services RAG application covering policies, reports, and metadata using Azure Data Lake Storage, Azure indexing, and Azure OpenAI; owned ingestion, preprocessing, chunking, embeddings, indexing, retrieval, and LLM integration.
• Developed and optimized the backend for a RAG-based 3D Avatar application for a senior leader, using prompt tuning and retrieval controls to keep responses grounded in the application's knowledge base; the backend was integrated with the Proto avatar experience by the relevant team.
• Built FastAPI-based backend services for LLM and agent workflows and used LangGraph to orchestrate multi-agent application flows; integrated AI agents with Databricks and Snowflake for enterprise data access and AI-driven querying and transformation.
• Performed RLHF-style LLM data annotation and response ranking to support LLM training and improve response quality; applied RAGAS and TruLens for RAG and LLM application evaluation.

Master of Business Administration in MBA – VIT (Distance Learning) (2026-01)

Bachelor of Computer Applications – Marian College Kuttikkanam (Autonomous) (2019-06 – 2022-03)