Senior AI/ML Engineer GenAI & MLOps Expert

LEELA MRUNALINI G.K.

Senior AI / Machine Learning Engineer

Alpharetta, GA Open to Hybrid

Email: ****@*********.*** LinkedIn: https://www.linkedin.com/in/leela-mrunalini-gk-a92ba723/ Phone: +1-470-***-**** GitHub: https://github.com/leelaMrunalini

PROFESSIONAL SUMMARY

Senior AI / Machine Learning Engineer with 8+ years of experience building and deploying scalable ML and Generative AI solutions across finance and healthcare domains.

Strong expertise in end-to-end machine learning pipelines, including data ingestion, feature engineering, model training, validation, deployment, and production monitoring.

Proven experience in statistical modeling, supervised and unsupervised learning, anomaly detection, time series analysis, and feature engineering.

Advanced skills in NLP and Deep Learning, including Transformers, BERT, GPT, LLMs, text classification, and prompt engineering.

Hands-on experience developing Generative AI applications, including Retrieval-Augmented Generation (RAG), vector databases (FAISS), and LangChain-based solutions.

Extensive background in productionizing ML models, exposing them via RESTful APIs, and integrating with enterprise systems.

Strong experience with cloud-based ML platforms such as AWS (SageMaker, S3, Lambda) and Azure ML / Azure OpenAI.

Proficient in Python, SQL, ML frameworks (Scikit-learn, TensorFlow, PyTorch), and MLOps tools including Docker, MLflow, CI/CD.

Demonstrated ability to lead cross-functional teams, mentor engineers, and deliver AI solutions aligned with business and regulatory requirements.

CORE SKILLS:

Machine Learning & Statistics: Supervised/Unsupervised ML, Advanced Statistical Modeling, Time Series, Anomaly Detection, Feature Engineering

Deep Learning & NLP: Transformers, LLMs, BERT, GPT, Text Classification, NER, Topic Modeling, Prompt Engineering

Generative AI: RAG pipelines, Vector Databases (FAISS), LangChain, Lang Graph, Agentic AI

MLOps & Deployment: Model Deployment, CI/CD, MLflow, Docker, REST APIs, Monitoring

Cloud Platforms: AWS (SageMaker, S3, Lambda), Azure ML, Azure OpenAI

Programming & Data: Python, SQL, Pandas, NumPy, Scikit-learn, TensorFlow, PyTorch

Visualization & Apps: Streamlit, Dash, Matplotlib, Seaborn

Tools: Git, Hugging Face, Databricks

PROFESSIONAL EXPERIENCE:

Senior Data Scientist / AI-ML Engineer

ACI Worldwide – Alpharetta, GA

2023 – Present

Architected and led the development of enterprise-scale, real-time machine learning systems for payment fraud detection, handling millions of transactions daily with low-latency and high-availability requirements.

Designed and implemented advanced statistical and ML models (Logistic Regression, XGBoost, Random Forest, Deep Neural Networks) to detect anomalous and fraudulent transaction patterns, significantly improving precision-recall trade-offs in high-risk scenarios.

Built robust end-to-end ML pipelines including data ingestion, feature engineering, model training, hyperparameter tuning, validation, deployment, and performance monitoring in production.

Developed LLM-powered analyst-assist and explainability solutions using Retrieval-Augmented Generation (RAG) with LangChain, FAISS, and Azure OpenAI, enabling fraud analysts to query transaction insights and model rationale in natural language.

Integrated machine learning models with RESTful APIs to enable seamless consumption by downstream payment platforms and internal risk systems.

Deployed and managed models on Azure ML, leveraging scalable compute, automated pipelines, and model versioning to ensure reliability and compliance.

Collaborated closely with data engineering, platform, and product teams to align AI solutions with business and regulatory requirements.

Provided technical leadership and mentorship to junior engineers, driving best practices in ML engineering, code quality, model governance, and production readiness.

Data Scientist / Machine Learning Engineer

Optum (UnitedHealth Group) – USA

2021 – 2023

Led the design and development of an enterprise-scale Healthcare Prior Authorization Optimization platform aimed at reducing manual review workload and improving approval turnaround times.

Analyzed large volumes of clinical notes, physician documentation, medical records, and authorization requests to identify automation opportunities using NLP and ML techniques.

Ensured compliance with HIPAA regulations, data privacy standards, and enterprise security guidelines.

Developed RESTful APIs to deploy ML models and enable real-time integration with enterprise healthcare systems.

Designed end-to-end Natural Language Processing pipelines for text ingestion, cleansing, normalization, tokenization, and feature extraction from unstructured healthcare data.

Implemented multiple text representation techniques including TF-IDF, Bag-of-Words, word embeddings, and contextual embeddings to improve document understanding.

Built and evaluated machine learning models such as Logistic Regression, Support Vector Machines (SVM), Random Forests, and Deep Learning architectures for document classification.

Developed models to automatically classify authorization requests into approval, denial, and review-required categories.

Applied advanced feature engineering using diagnosis codes (ICD), procedure codes (CPT), provider history, and utilization metrics.

Performed extensive Exploratory Data Analysis (EDA) and statistical modeling to identify key drivers impacting authorization decisions.

Reduced manual intervention by routing high-confidence approvals and flagging complex cases for clinical review.

Data Scientist

TCS – Bangalore, India

2017 – 2021

Delivered multiple end-to-end analytics and machine learning projects for global enterprise clients across retail, telecom, BFSI, and operations domains.

Worked directly with business stakeholders to translate complex business problems into structured analytical and ML use cases.

Designed scalable data ingestion and preprocessing pipelines for structured and semi-structured datasets from multiple sources.

Conducted comprehensive data quality assessments, handling missing values, outliers, and inconsistencies.

Applied statistical analysis and hypothesis testing to validate business assumptions and model insights.

Implemented reusable ML workflows using Python, SQL, and enterprise data platforms.

Deployed models into production environments and supported post-deployment monitoring and enhancements.

KEY PROJECTS

Generative AI – Domain-Specific Virtual Assistant (RAG)

Designed and implemented a Retrieval-Augmented Generation (RAG) system using FAISS, LangChain, and GPT-based LLMs.

Enabled ingestion of enterprise PDFs, chunking, embedding generation, semantic retrieval, and grounded answer generation.

Deployed using Streamlit with secure API integration.

Fraud Detection ML Platform

Built scalable ML pipelines for real-time fraud detection with explainability.

Integrated models with REST APIs and deployed on cloud infrastructure.

EDUCATION

Master of Technology (M.Tech) – Electrical Power Systems

Master Certification – Data Science & Deep Learning (GUVI)

CERTIFICATIONS

Generative AI with LLMs – Coursera (DeepLearning.AI)

AWS Machine Learning Specialty – In Progress

LEADERSHIP & SOFT SKILLS

Technical Leadership & Mentoring

Strategic AI Roadmap Development

Cross-functional Collaboration

Strong Communication & Problem Solving

Contact this candidate