Skip to content

Pallavi Priya Kadiyam

Applied AI Engineer • Dallas, TX • p****************@gmail.com • 945****605 • linkedin.com/•••••

Professional Summary

Applied AI Engineer with 5+ years designing and delivering production-grade Generative AI, RAG, and agentic AI systems for telecom and healthcare. Experienced building enterprise LLM platforms, semantic search and retrieval pipelines, multi-agent orchestration, evaluation frameworks, and scalable cloud-native AI services on AWS and GCP. Proven track record reducing search latency and incident resolution times, improving retrieval relevance, and operationalizing responsible AI for regulated environments.

Technical Skills

Frameworks and Libraries: LangChain
Cloud and DevOps: AWS Bedrock,AWS SageMaker,AWS Lambda,AWS ECS,Google Vertex AI,Kubernetes,Terraform,CI,CD
Generative AI & LLMs: LangGraph,Prompt Engineering,Fine-tuning,LLOps,Agentic Workflows
Retrieval & Semantic Search: Retrieval-Augmented Generation RAG,Semantic Search,Vector Search,Embeddings,Hybrid Retrieval,Context-aware Chunking
Observability & MLOps: Langfuse,Prometheus,Grafana,OpenTelemetry

Work Experience

Verizon
Dallas, TX
AI Engineer — Agentic Systems & LLM Platform
Jan 2024 – Present
Tech Stack: OpenSearch, LangChain, LangGraph, FastAPI, AWS Bedrock, S3, Prometheus, Grafana, OpenTelemetry, Langfuse
  • Designed and deployed enterprise RAG solutions across three business units using OpenSearch, vector indexing, hybrid retrieval, and context-aware chunking, reducing semantic search latency by 65% and accelerating access to critical knowledge.
  • Built the enterprise's first production semantic search pipeline with vector embeddings and re-ranking strategies to improve search relevance and accelerate time-to-answer for internal users.
  • Developed LangChain and LangGraph-based agentic workflows to automate multi-step business processes, orchestrate enterprise tools, and maintain conversational state for reliable AI-driven operations.
  • Engineered evaluation frameworks, golden datasets, and validation workflows to measure groundedness, retrieval accuracy, and model reliability for production GenAI applications.
  • Implemented production observability and incident monitoring using Prometheus, Grafana, OpenTelemetry, and Langfuse to track latency, token usage, and retrieval performance, cutting incident resolution time by 50%.
  • Developed and deployed FastAPI-based AI services integrated with internal REST APIs and ERP systems on AWS, achieving 99.5%+ uptime and reducing infrastructure costs by 20% through cloud optimization.
Trillium Health Resources
Greenville, NC
Data Engineer — AI & GenAI Platforms
Apr 2023 – Dec 2023
Tech Stack: Google Vertex AI, LangChain, OpenAI APIs, GCP IAM, BigQuery, Langfuse
  • Built semantic search and RAG pipelines over HIPAA-regulated clinical document repositories on Google Vertex AI using LangChain and OpenAI APIs, improving clinical information discovery by 60%.
  • Designed prompt engineering strategies and evaluation frameworks tailored for clinical use, improving response quality, groundedness, and retrieval effectiveness for patient- and clinician-facing applications.
  • Developed and deployed real-time inference services for anomaly detection and patient risk scoring, moving four AI use cases from prototype to production in a regulated healthcare environment.
  • Established AI observability and automated evaluation frameworks with 500+ benchmark test cases to improve model reliability and stability across production releases.
  • Collaborated with clinical stakeholders to translate business requirements into compliant AI solutions and produced implementation guides and operational runbooks for cross-functional teams.
  • Implemented secure data handling, privacy controls, and access governance for clinical pipelines to maintain HIPAA compliance during model inference and data retrieval.
L&T Infotech
Hyderabad, India
Software Engineer — Backend, Java & Data Automation
Feb 2020 – Dec 2022
Tech Stack: Python, Java, PostgreSQL, MS SQL Server, REST APIs, Git, Airflow
  • Implemented Python and Java backend automation workflows and REST API integrations across high-volume enterprise pipelines, maintaining Git version control and architectural decision records.
  • Optimized database architecture, indexes, and execution plans across PostgreSQL and MS SQL Server to deliver a 40% improvement in reporting and query performance for enterprise dashboards.
  • Authored 100+ automated unit and integration tests to improve code coverage and reduce production defects, contributing to more reliable release cycles.
  • Designed and maintained ETL and data automation jobs to feed analytics and model feature pipelines, improving data availability for downstream AI use cases.
  • Applied secure development practices, performed code reviews, and worked in Agile delivery to ensure maintainable and secure software deployments.
  • Produced technical documentation, runbooks, and deployment playbooks to standardize onboarding and improve operational maintainability.
Agentic Systems

Projects

Enterprise Semantic Search & Document Intelligence Platform
Tools Used: LangChain, OpenSearch, Amazon Bedrock, OpenAI, FAISS, Pinecone, Langfuse, Pytest
  • Built a production-grade semantic search and RAG platform using vector embeddings, hybrid retrieval, and re-ranking to deliver citation-backed responses and reduce manual search effort by 60%.
  • Implemented prompt templates, responsible AI guardrails, and an evaluation pipeline with regression tracking and offline/online metrics to improve answer groundedness and user trust.
  • Integrated monitoring and automated tests to maintain retrieval quality and track model drift during deployments.
Agentic Workflow Platform — Content Intelligence & Automation
Tools Used: LangGraph, LangChain, Amazon Bedrock, OpenAI, Anthropic, OpenSearch, FastAPI, Prometheus, Grafana
  • Developed a multi-agent orchestration framework that integrated enterprise knowledge sources, operational APIs, and hybrid retrieval, reducing escalations and improving operational efficiency.
  • Implemented AI governance controls, conversational state management, and service-level observability to ensure reliable, auditable automation for enterprise workflows.
LLM Evaluation & Search Relevance Framework
Tools Used: LangChain, OpenSearch, Python, Pytest, Langfuse
  • Built an end-to-end GenAI evaluation framework leveraging golden datasets, A/B testing, and relevance scoring to measure retrieval quality and drive pipeline optimizations.
  • Delivered regression tracking and monitoring that enabled three pipeline improvements which increased answer groundedness and reduced incident resolution time.

Education

Texas A&M University Commerce
M.S., Business Analytics • Commerce, TX
Sreenidhi Institute of Science & Technology
B.Tech., Information Technology • Hyderabad, India

Certifications

AWS Certified Data Engineer – Associate — Amazon Web Services
Microsoft Certified: Azure Data Engineer Associate — Microsoft
Google Cloud Professional Data Engineer — Google Cloud

Powered by Drivetube · Create your own profile at drivetube.ai