Naganjali Pujitha
Data Scientist Engineer / Generative AI & LLM Prompts Engineer | Lang Chain | Machine Learning Engineer | MLOps Engineer | Analytics Engineer |
- Role
- Data Scientist at Wells Fargo
- Location
- New York, NY, US
- LinkedIn followers
- 500 followers
About Naganjali Pujitha
Naganjali Pujitha – Data Scientist | Data Engineer | Cloud & ML Enthusiast Tech…
Experience
Data Scientist
Dec 2024 — Present · New York, NY, US
Worked on POC of Model Context Protocol (MCP) to enhance the integration of diverse machine learning models within multi-agent AI systems, ensuring consistent model behavior and context-aware decision-making across retail and financial applications.• Worked on POC of workflow automation using n8n, optimizing complex data processes across cloud platforms and enhancing operational workflows, significantly reducing manual intervention and improving system integration for seamless data flow.• Worked POC of AI-driven workflows using CrewAI, streamlining multi-agent systems for retail intelligence, improving operational efficiency, and enhancing customer experience through dynamic AI interactions.• Integrated OCR tools (Tesseract, Tex tract) to digitize scanned financial documents, enhancing processing speed by 40%.• Engineered prompts and workflows for LLMs to generate high-quality text outputs aligned with compliance standards.• Developed multi-agent systems using Microsoft Autogen for summarization, financial analysis, and fact-checking.• Delivered production-ready ML models via AWS SageMaker to enhance credit risk scoring and customer segmentation.• Designed an Agentic RAG pipeline integrating FAISS + LangChain agents to perform contextual reasoning over 100K+ internal documents, boosting answer accuracy by 40%.• Built a multi-agent system using Microsoft AutoGen to handle tasks like document summarization, financial forecasting, and hallucination detection, enabling modular orchestration of GenAI workflows.• Implemented LangGraph to manage long-running agent workflows with stateful memory and retries, ensuring robustness in complex LLM-based decision flows.• Integrated LangGraph with prompt tracing, enabling explainable step-by-step flows across multi-agent collaboration tasks.
Education
Tirumala Engineering College(JNTUK)
Undergraduate, Electronics and Communication Engineering
Pace University - Seidenberg School of Computer Science and Information Systems
Master, Computer Science
Find verified contacts for anyone on LinkedIn
Unifers gives sales teams verified emails and direct dials, enriched profiles, and outreach that lands in the inbox.
Free plan included · No credit card required
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.