AI Engineer - Generative AI & LLMs Specialist

Naga Venkata Padala

Email: ***************@*****.***

Mobile: 214-***-****

LinkedIn: https://www.linkedin.com/in/pavan-papi-reddy-padala/ AI Engineer

PROFESSIONAL SUMMARY

AI/ML Engineer with 4+ years of experience designing and deploying Generative AI workflows, RAG systems, and recommendation engines across finance, retail, and cloud environments.

Expertise in orchestrating LLMs and Foundation Models on AWS Bedrock for regulatory compliance, structured data extraction, and automated reporting in banking and enterprise applications.

Designed and implemented autonomous agentic workflows that summarize transaction patterns, generate narrative risk reports, and reduce manual investigation time for compliance and fraud teams.

Built and optimized high throughput streaming pipelines using Spark Structured Streaming and low-latency feature stores on distributed NoSQL systems to enable real-time fraud detection and behavioral inference.

Presented technical work clearly to technical and non-technical stakeholders, enhancing project understanding and collaboration.

Contributed to rebuilding a cohesive team by mentoring junior members, resulting in improved team performance.

Supported cross-functional teams and partnered with leadership for adoption strategy, aligning goals and boosting productivity.

TECHNICAL SKILLS

Generative AI & LLMs - Large Language Models (OpenAI, Gemini, Claude), AWS Bedrock, LangChain, LangGraph, RAG Architectures, Vector Databases (FAISS, Pinecone), Agents (MCP, AutoGen), Prompt Engineering, Hugging Face Transformers.

Machine Learning & Deep Learning - PyTorch, TensorFlow, Scikit-learn, Pandas, NumPy, XGBoost, Computer Vision (CNNs, OpenCV, YOLO/OmniParser), Time Series (Prophet, LSTM), Explainable AI (SHAP, LIME)., AI/ML application development, AI-enabled applications, AI Infrastructure platform, AI Engineers, AI tools, AI/ML platform engineering

Cloud Platforms - AWS (SageMaker, Bedrock, Lambda, Glue, EMR, Kinesis), Azure (Azure ML, Databricks, AKS, Synapse), Serverless Computing., Azure OpenAI, AWS AI/ML

MLOps & DevOps - Docker, Kubernetes (EKS/AKS), Terraform (IaC), MLflow, Apache Airflow, CI/CD

(GitHub Actions, Azure DevOps), Git.

Big Data & Engineering - Apache Spark (PySpark), Kafka, Databricks, SQL, NoSQL (DynamoDB, MongoDB), Data Lakes (S3, ADLS), ETL Pipelines.

Programming Languages - Python (Advanced), R, Scala, SQL, Bash., .py (Python)

Visualization & Deployment - FastAPI, Flask, Streamlit, Power BI, Grafana, Prometheus.

NLP Tools & Libraries - SpaCy, NLTK, TextBlob, BERT, RoBERTa

AI Tools & Platforms - AI Foundry, Copilot Studio, Microsoft Copilot

Software Architecture - reference architectures

PROFESSIONAL EXPERIENCE

U.S. Bank Jan 2024 – Present

AI Software Engineer

Developed Generative AI workflows using AWS Bedrock, orchestrating Foundation Models to automate regulatory compliance checks and extract structured insights from complex banking documents.

Designed autonomous agentic workflows to summarize transaction patterns and generate narrative risk reports, significantly reducing the manual investigation time for fraud analysts.

Implemented LLM evaluation frameworks, hallucination detection, and PII redaction layers to ensure generated outputs strictly adhered to financial data privacy and security standards.

Architected a high-throughput Spark Structured Streaming pipeline for Instant Payments fraud detection, enabling sub-second risk analysis on live transaction flows.

Engineered a low-latency offline-to-online feature store backed by distributed NoSQL systems to provide real-time behavioral signals during model inference.

Led AI Platform Development and AI Infrastructure platform initiatives, enhancing processing efficiency by 40% and supporting scalable AI-enabled applications across enterprise environments.

Spearheaded Application Engineering and Integration & Security projects, ensuring compliance with security compliance standards and optimizing integration processes for improved system reliability.

Facilitated Team & Collaboration efforts by mentoring AI Engineers, contributing to rebuilding cohesive teams and advancing AI/ML platform engineering best practices. Target July 2023 – Dec 2023

Machine Learning Engineer

Built an end-to-end RAG system on AWS to handle complex internal queries, managing the full pipeline from document ingestion and vectorization to final LLM response generation.

Improved retrieval accuracy by implementing a hybrid search strategy that combines standard keyword search with vector-based semantic search, adding a re-ranking step to ensure the most relevant context reaches the model.

Set up a concrete testing framework using RAGAS metrics (Faithfulness, Answer Relevance) and manually verified golden datasets to tune chunk sizes and prompt templates based on actual performance data.

Solved context window limitations by implementing parent document retrieval and metadata filtering, ensuring the model accessed the full scope of large documents without losing specific details.

Developed a Two-Tower Neural Network recommendation system on AWS SageMaker, optimizing the model to balance high click-through rates with actual conversion value.

Architected enterprise AI platforms from ground up, leveraging AI Foundry and Copilot Studio, resulting in a 50% increase in deployment speed for AI tools.

Utilized Azure OpenAI and AWS AI/ML services to deliver cloud-based AI solutions, reducing operational costs by 25% and accelerating time-to-market for AI applications.

Enhanced enterprise AI integration and secure system design, implementing enterprise identity management and identity security integration for robust security compliance. Infosys Jan 2021 – Oct 2022

Data Scientist

Built predictive churn models using Logistic Regression and Random Forests, tuning them to find the right balance between precision and recall so we could accurately flag at-risk customers.

Deployed these models as serverless REST APIs using Flask and AWS Lambda, allowing the web application to fetch real-time risk scores for active users instantly.

Implemented automated data drift and schema monitoring using TensorFlow Data Validation (TFDV), tracking feature distribution changes to ensure we caught data issues before they broke the model.

Analyzed customer behavior using K-means clustering and NLP on review data, which helped the marketing team move away from generic blasts to targeted, personalized campaigns.

Automated the retraining pipeline using AWS Step Functions, setting up triggers that pulled fresh data from Redshift and updated the models without anyone needing to run manual scripts.

Implemented reference architectures and standardized core components, improving AI/ML application development efficiency by 30% in large enterprise consulting environments.

Developed AI-enabled applications for the oil & gas energy sector, driving a 20% increase in ROI through innovative AI tools and enterprise AI integration.

Presented technical work clearly to technical and non-technical stakeholders, supporting cross-functional teams and partner leadership in adoption strategy and secure system design. CERTIFICATIONS

AWS Certified AI Practitioner – Amazon Web Services

AWS Certified Machine Learning Engineer Associate - Amazon Web Services

TensorFlow Developer Certificate – TensorFlow / Google

Hugging Face AI Agent Fundamentals - Hugging Face EDUCATION

Master of Science in Data Science - University of Massachusetts Dartmouth

Bachelor of Technology in Mechanical Engineering - Amrita University

Contact this candidate