Sujan Reddy

About Sujan Reddy

GenAI Engineer & Data Engineer with 5+ years building retrieval systems, LLM-powered document understanding, and secure production pipelines across healthcare, finance, and enterprise settings. I design RAG architectures (Pinecone/ChromaDB), integrate LangChain, and harden solutions with prompt templates, PHI masking, and auditability to meet HIPAA/SOC-2 requirements.I’ve shipped FastAPI services, containerized with Docker/Kubernetes (AKS/EKS), orchestrated Airflow DAGs, and standardized ML workflows using MLflow, Git-driven experiment lineage, and CI/CD. Prior work spans OCR (Tesseract), NLP (spaCy, cTAKES, ClinicalBERT), and scalable data foundations on AWS/GCP/Azure.I love turning messy text and transactional signals into reliable, explainable services—then instrumenting them for latency, drift, and quality so teams can move faster with confidence.Focus: RAG systems • LLMOps • secure GenAI • fraud/claims analytics • production AI platforms.

Experience

1. Generative Ai Engineer

   Annalect

   Jan 2024 — Present

   Built RAG pipelines: token-aware chunking, embeddings, and cosine similarity retrieval (Pinecone/ChromaDB).Controlled output style via role-based prompt templates; enforced brand/PII safety and per-agency isolation.Served cached answers for high-similarity queries; logged prompt/source/model metadata for audits & tuning.Exposed FastAPI endpoints for summaries, target audiences, and campaign suggestions; integrated results into Tableau.Migrated/validated prompts for AWS Bedrock; balanced latency vs. quality using response metrics & routing.

Education

• Indiana Institute of Technology

  Master of Science - MS

  2022 — 2023