agentic software engineer

Build secure enterprise GenAI and agentic-AI capabilities for banking operations, risk, and technology teams, owning RAG design, agent orchestration, evaluation, production reliability, and cloud deployment across high-governance workflows.
Designed multi-agent investigation workflows that decompose complex requests, gather evidence, compare findings, and produce cited recommendations for human review.
• Built enterprise RAG across policies, procedures, regulatory guidance, historical cases, architecture documents, and ownership metadata.
• Engineered hybrid retrieval using dense embeddings, keyword search, metadata filters, query rewriting, evidence-aware reranking, and context deduplication.
• Implemented governed LLM tool calling for case search, policy retrieval, workflow status, service telemetry, and approved actions with scoped permissions and audit records.
• Established AI evaluation covering retrieval recall, groundedness, case-resolution accuracy, task completion, hallucinations, latency, and cost.
• Improved grounded-answer accuracy by 24% and reduced end-to-end latency by 31% through chunking, reranking, citation validation, parallel tool execution, caching, streaming, and model routing.
• Added prompt-injection screening, PII controls, least-privilege actions, human approval gates, deterministic validation, fallbacks, and output-policy checks.
• Deployed autoscaled AI services on Kubernetes and AWS with OpenTelemetry tracing, quality dashboards, spend alerts, safe rollouts, and automated rollback.

Built production insurance ML and document-intelligence services that transformed forms, correspondence, claim events, and policy data into structured features, classifications, prioritization signals, and recommendation workflows.
• Built Kafka and Spark pipelines to ingest claim events, scanned documents, correspondence, policy metadata, and workflow outcomes into model-ready features.
• Developed document extraction and classification models for form type, coverage attributes, claim intent, missing information, and processing priority.
• Trained NLP classifiers for topic, urgency, sentiment, and routing of adjuster notes and customer correspondence.
• Created recommendation and ranking services that prioritized processing queues and surfaced related policies, prior cases, and likely next steps.
• Improved document-routing precision by 19% through layout-aware features, hard-negative sampling, calibrated thresholds, and error analysis with claims specialists.
• Standardized experimentation, model registry, approvals, lineage, reproducibility, and rollback using MLflow.
• Delivered low-latency model-serving APIs with FastAPI and gRPC, Redis caching, batch inference, graceful fallback, and Kubernetes-based canary deployment.
• Implemented monitoring for drift, feature freshness, prediction distributions, latency, errors, and downstream outcome quality.

Modernized Prudential policy-servicing and claims integration across legacy and cloud systems, owning APIs, event-driven services, AWS migration, resilience patterns, performance optimization, and operational tooling.
• Owned Java and Spring Boot APIs for policy lookup, customer updates, document status, claim intake, and servicing workflow actions.
• Designed event-driven integration services with Amazon Kinesis, SQS, and SNS to process policy and claim events across operational systems.
• Integrated rules and risk-scoring outputs into low-latency, versioned APIs with validation, explainability metadata, and fallback logic.
• Reduced p95 API latency by 37% through Redis caching, DynamoDB access-pattern redesign, request coalescing, and parallel downstream calls.
• Migrated services to AWS using ECS, Lambda, API Gateway, RDS, DynamoDB, S3, CloudWatch, and Terraform-managed infrastructure.
• Improved resilience with idempotent consumers, circuit breakers, bounded retries, dead-letter recovery, health checks, and operational runbooks.
• Built React and TypeScript dashboards for claim intake, document exceptions, processing lag, service health, and operational workload.
• Expanded unit, integration, consumer-contract, load, and failure-recovery testing in CI/CD and contributed to architecture and production-readiness reviews.

orithms & Data Structures, Database Systems, Software Engineering, Computer Networks, Operating Systems, Artificial Intelligence, Machine Learning, Distributed Systems

Academic Project / Thesis:
Intelligent Document Classification and Information Retrieval — Designed a machine-learning-based system for classifying documents and retrieving relevant information using NLP techniques, with emphasis on feature extraction, ranking, and evaluation.