Agent Orchestration & LLM Integration • Developed, integrated, and evaluated four model configurations of the new AWS Claude Bedrock model against the baseline Azure OpenAI models inside the agent orchestration platform, using LiteLLM as an extension to the OpenAI Agents SDK. • Built a Jenkins regression pipeline triggered on PR that uploads manifest/agent-card changes to blob storage, provisions an ephemeral environment for AO, and runs the full agent regression suite. • Prototyped cloud-native AI governance (Bedrock Guardrails + Azure AI Content Safety) for pre-inference prompt checking and sanitization. • Contributed to the manifest/agent-card configuration store service (REST API, CI/CD checks, multi-region/multi-cloud deploys). • Built a blob migration tool for migrating already cloud-stored manifests on schema changes. Cloud Infrastructure & Deployments (AWS + Azure / Kubernetes) • Deployed and operated microservices on AWS and Azure Kubernetes with Argo CD and Helm using path-filtered pipelines; oversaw multi-region rollouts, scaling, monitoring (Prometheus/Grafana), and incident response. • Built path-filtered CI checks per service with Kafka Testcontainers for realistic integration tests and live Alembic migrations against PostgreSQL.

Document Ingestion Platform • Contributed to re-architecting the ingestion system from a single synchronous application into a three-stage asynchronous Kafka pipeline (ingestion → chunking → embedding services), enabling reliable large-file handling and independent horizontal scaling. • Designed REST APIs to ingest and index multiple document types (PDF, Markdown, plaintext), with system- and user-level collection permissions, tenant isolation, and per-service code metrics. • Designed and built a cloud-agnostic storage handler library that transparently routes documents to Azure Blob Storage, Azure File Storage, or AWS S3 based on the deployed cloud — later reused as the foundation for storing agent manifests/cards. • Implemented Kafka/PostgreSQL health-check monitoring wired into Kubernetes readiness probes, async end-to-end tests, and document retention/delete policies across Azure and S3. • Led comparative POCs of vector databases (Pinecone, Milvus, pgvector) for storing document embeddings, assessing query performance, scalability, and operational overhead. • Built multiple Jenkins pipelines to publish shared Python libraries to internal package registries, with automated version-bump on release.

Roomio: Improved the project to meet client requirements and enhance its appeal for future clients. Trax: Migrated a legacy application to a new framework and built new features, collaborating closely with the client to align with their specific requirements.

Designed and developed a web application that accurately calculates the effort required to construct IT projects based on the specific requirements provided.

Relevant Coursework: Artificial Intelligence, Web Technologies, Adaptive Systems, Multi-Agent Systems, Semantic Web, Distributed Processing, Information Security, Information Retrieval, Machine Learning, Data Distribution, Web Application Development, Social Network Analysis, Interactive Systems.

Relevant Coursework: Data Structures, Algorithm Design, Analysis of Algorithms, Programming Paradigms, Object-Oriented Programming (OOP), Computers Programming, Operating Systems Usage, The Architecture of Computing Systems, Software Systems for Computer Networks, Parallel and Distributed Algorithms.