Sweta Pati's Personal Website

spati2203@gmail.com · Hilliard, OH · Open to relocate

prof_pic_close.png

ML Engineer @ Central Mutual Insurance

Resume (PDF)

Hello! I’m Sweta Pati, an AI/ML Engineer with 5+ years of experience building production Machine Learning, NLP, Generative AI, and Agentic AI systems for Fortune 500 clients across insurance, recruitment, fintech, and academic research.

Experience

  • ML Engineer, Central Mutual Insurance (Jan 2026 – Present): Architecting production AI systems for insurance workflows. Shipped a multi-agent Claims Leakage platform (LangChain Deep Agents on LangGraph, 8 concurrent specialist agents) that cut manual claim-review effort by 90%, with full observability via OpenTelemetry and Azure Application Insights, and built an LLM-powered document comparison pipeline (Azure Document Intelligence, GPT-4o) that cut 50–300 page rate-manual reviews from 5 days to 30 minutes across 19 lines of business.

  • Machine Learning Engineer, 4A Consulting (Jun 2025 – Jan 2026): Built AI-driven recruitment automation: an LLM-based interview scoring pipeline that improved consistency by 30% across 100+ assessments, and a LangChain/FAISS RAG system handling 4,000+ monthly queries.

  • NLP Graduate Research Assistant, George Mason University (May 2024 – Jan 2025): Co-authored Instruction-Tuning LLMs for Event Extraction with Annotation Guidelines, accepted at ACL 2025 (Findings); achieved +10% Trigger / +5% Argument F1 gains on ACE05 and RichERE by fine-tuning LLaMA-3.1/3.2 and Qwen2.5-Coder with LoRA and schema-guided instruction tuning.

  • Software Engineer II (Data Science), HighRadius Technologies (Jul 2019 – Aug 2023): Led a 10-member team delivering ML solutions for 15+ Fortune 500 clients (Kraft Heinz, Clorox, L’Oréal, Ferrero, Duracell), improving dollar recovery by 12% with fraud-detection models, on-time collections by 17%, and cutting ML deployment time by 40% via automated pipelines on AWS.

Technical Expertise

  • LLMs & Fine-Tuning: Transformers, LoRA/QLoRA, PEFT, RLHF, quantization, instruction tuning, prompt engineering.
  • Agentic AI & RAG: LangChain, LangGraph, Deep Agents, MCP tools, agent evaluation & observability, retrieval-augmented generation.
  • Machine Learning & NLP: Classification, regression, time-series forecasting, event extraction: applied across insurance, recruitment, and fintech.
  • MLOps & Deployment: Docker, FastAPI, Azure (Document Intelligence, DevOps CI/CD, App Insights, Foundry), Kubernetes, MLflow, Weights & Biases.

Education

  • M.S., Computer Science (GPA 3.93/4), George Mason University, 2023–2025
  • B.S., Computer Science, KIIT University, 2016–2020
  • Credentials: AI Agents Fundamentals (Hugging Face) · Google AI Essentials · Fine-Tuning Large Language Models

Always happy to connect; reach out via email or the links below.

News

Jan 12, 2026 Excited to share that I’ve joined Central Mutual Insurance as an ML Engineer and relocated to Hilliard, Ohio! Looking forward to architecting production AI systems (including multi-agent platforms and LLM-powered document automation) for insurance workflows.
Jun 30, 2025 Excited to share that I’ve joined 4A Consulting, LLC as an AI/ML Engineer, where I’m building modular LLM evaluation systems and GenAI pipelines for recruitment platforms serving Fortune 500 clients.
May 15, 2025 🎉 Our paper, Instruction-Tuning LLMs for Event Extraction with Annotation Guidelines has been accepted to ACL 2025 (Findings)! This work is co-authored with Saurabh Srivastava and Dr. Ziyu Yao at George Mason University – College of Engineering and Computing. Grateful to have worked under the guidance of Prof. Ziyu Yao at GMU’s NLP Lab.
May 05, 2025 Honored to receive the Distinguished Academic Achievement Award from the Department of Computer Science at George Mason University!
Apr 15, 2025 Participated in Bitcamp Hackathon organized by University of Maryland, College Park and built TechMate - Turning “Tech Trouble” into “Tech Triumph” Together. TechMate ia a voice-enabled AI assistant designed to help seniors navigate through any technology challenges with confidence.
Feb 15, 2025 I’m excited to share that we have submitted our research titled Instruction-Tuning LLMs for Event Extraction with Annotation Guidelines to ACL 2025 under the guidance of Prof. Ziyu Yao.
Jun 01, 2024 I’m excited to announce that I am starting a Research Assistant position under the guidance of Prof. Ziyu Yao in GMU’s NLP Lab! I will be working in the domain of Event Extraction using instruction-tuned large language models (LLMs).
Mar 01, 2024 Excited to share that I’ve joined The Edge, Mason Recreation as a Team Building Facilitator (Graduate Professional Assistant) and I’m looking forward to contribute to creating impactful experiences while further developing my leadership and communication skills!
Aug 21, 2023 I’m thrilled to share that I’ve begun my Master’s in Computer Science, specializing in Machine Learning at George Mason University!