Clinton Brown

Passionate and results-driven Full Stack Engineer with over 8 years of experience architecting and scaling high-performance distributed systems, production-grade AI platforms, and cloud-native applications. Specialized in designing end-to-end Retrieval-Augmented Generation (RAG) systems, including large-scale document ingestion, semantic chunking, embedding pipelines, dense and hybrid retrieval, reranking, and vector search using technologies such as Microsoft Azure Cognitive Search, Milvus, and FAISS-based indexes. Strong expertise in building low-latency, high-concurrency APIs and microservices using Python, Node.js, FastAPI, asyncio, REST, and gRPC, consistently supporting strict p95/p99 latency and throughput requirements.

Proven track record of developing event-driven architectures on Azure using Azure Functions, Service Bus, AKS, and Event Grid to support both real-time and batch processing workloads. Hands-on experience integrating and deploying LLM-powered applications in production through Azure OpenAI, LangChain, prompt orchestration, token-budget control, and evaluation frameworks using retrieval, generation, and task-specific metrics. Deep expertise in observability, fault-tolerant system design, idempotent processing, retries, DLQs, rate limiting, and secure backend architectures using RBAC, OAuth2/OIDC, JWT, and TLS/mTLS.

Experienced across Azure cloud ecosystems, with a strong focus on building scalable, reliable, and intelligent backend systems across AI, fintech, healthcare, ad-tech, and enterprise domains.

Senior Full Stack Engineer (AI & Backend Systems) - SmartMetrics - Chattanooga, TN

(2024-09 - 2026-03)

• Architected and built a scalable RAG-based platform capable of processing and retrieving multi-format data, including PDF, HTML, and images, to support intelligent search and downstream LLM applications.
• Designed and implemented a high-throughput ingestion pipeline using asynchronous and event-driven processing patterns to efficiently handle large-scale document workloads.
• Built FastAPI-based backend services and developed lightweight frontend dashboards for search, ingestion monitoring, and system observability (Angular 18+ / TypeScript).
• Integrated the Unstructured framework to perform document parsing, layout-aware chunking, and metadata extraction, significantly improving semantic quality and usability of processed data.
• Stored processed document data and metadata in Azure Blob Storage to ensure durability, scalability, and reliable access across distributed services.
• Designed a fully decoupled event-driven architecture using Azure Service Bus and Event Grid to trigger backend services and coordinate data processing workflows.
• Built resilient ingestion and processing workflows by implementing retry strategies, dead-letter queue handling, and fault-tolerant mechanisms to ensure reliable data processing under failure conditions.
• Developed a modular and extensible embedding service using a registry and adapter design pattern to support multiple embedding providers and models.
• Integrated multiple foundation models through Azure OpenAI, enabling flexible model selection based on performance and cost considerations.
• Generated high-quality vector embeddings for document chunks and indexed them into Azure Cognitive Search using k-NN vector search to enable scalable semantic retrieval.
• Designed embedding pipelines that support both batch processing for large datasets and incremental updates for continuously incoming data streams.
• Built a high-performance semantic search API using FastAPI, optimized for low-latency responses and capable of handling high-concurrency query workloads.
• Implemented Max Marginal Relevance (MMR) within the retrieval pipeline to improve both the diversity and contextual relevance of returned results.
• Refactored backend and infrastructure architecture using Terraform and Azure DevOps to eliminate circular dependencies, enforcing a one-directional dependency flow from application services to Azure Functions and infrastructure components, resulting in clearer service boundaries and improved maintainability.
• Designed and implemented secure backend systems by applying RBAC and OAuth2/OIDC-based authentication and authorization mechanisms to protect services and data access.
• Designed, provisioned, and deployed cloud infrastructure using Terraform and Azure DevOps pipelines, containerized services with Docker on AKS, and established CI/CD pipelines with deployment monitoring.
• Implemented observability using OpenTelemetry with Azure Monitor and contributed to system design, architectural decisions, and code reviews across distributed components.
• Leveraged AI-assisted development platforms such as Cursor, Claude, and Lovable to accelerate backend service development, debugging, and rapid prototyping of RAG-based features and internal tools.

Software Engineer - AI/ML and Back-end - WellTech Analytics - Memphis, TN

(2022-04 - 2024-08)

• Architected and developed a scalable RAG-based healthcare text classification platform for patient messages, clinical notes, insurance documents, lab reports, and multi-format medical records.
• Built internal frontend tools and dashboards for healthcare workflow monitoring, model evaluation, and labeling workflows using Angular 18+ / TypeScript.
• Designed and implemented high-throughput healthcare document ingestion pipelines using asynchronous and event-driven patterns.
• Integrated Unstructured framework for medical document parsing, layout-aware chunking, OCR-enabled text extraction, and metadata enrichment.
• Built secure preprocessing workflows for healthcare text normalization, PHI-aware document handling, and metadata extraction.
• Generated vector embeddings for medical document chunks and indexed them into Azure Cognitive Search using k-NN and HNSW.
• Designed retrieval-augmented classification architecture for LLM-assisted healthcare inference.
• Implemented Max Marginal Relevance, retrieval ranking, and prompt engineering to improve relevance and reduce hallucinations.
• Integrated multiple foundation models through Azure OpenAI.
• Developed modular embedding and inference services using registry and adapter patterns.
• Built fine-tuned transformer-based healthcare classification models.
• Developed resilient event-driven workflows with Azure Blob Storage, Service Bus, Event Grid using retries, dead-letter queues, and fault-tolerant orchestration.
• Built low-latency FastAPI services for real-time healthcare classification, semantic retrieval, and system integration.
• Designed secure backend systems using RBAC and OAuth2/OIDC authorization.
• Implemented evaluation pipelines using accuracy, precision, recall, F1-score, confusion matrices, and misclassification tracking.
• Provisioned infrastructure with Terraform, deployed Dockerized microservices on AKS, and implemented Azure DevOps CI/CD pipelines with Slack notifications.
• Implemented observability with OpenTelemetry and Azure Monitor.
• Leveraged AI-assisted development tools including Cursor and Claude to accelerate RAG pipelines and internal tooling.

Full Stack Engineer - TargetIQ - Chattanooga, TN

(2017-12 - 2022-03)

• Architected and delivered low-latency, high-throughput Node.js microservices using Express.js, NestJS, and gRPC to power real-time bidding and serving pipelines, consistently achieving sub-50 ms p99 latency under high-QPS production traffic.
• Built internal operational web dashboards and admin tools for campaign management and bidder configuration using Angular 18+ / TypeScript.
• Owned end-to-end bidder and serving-path services, including OpenRTB bid-request parsing, eligibility filtering, dynamic pacing algorithms, frequency capping, budget enforcement, and auction decisioning across distributed systems.
• Engineered high-throughput Kafka streaming pipelines using Node.js producers and consumers to process billions of daily events.
• Implemented advanced Kafka consumer-group architectures with partition-aware scaling, manual offset management, key-partitioned event routing to preserve per-key ordering, replay and backfill pipelines, and consumer-lag monitoring to ensure deterministic and fault-tolerant event processing.
• Improved event-processing reliability using idempotent handlers, deduplication keys, exponential retry and backoff strategies, DLQ pipelines, and circuit-breaker patterns.
• Optimized streaming performance through micro-batching, compression techniques, asynchronous processing pipelines, backpressure handling, and fine-grained concurrency tuning.
• Designed and implemented production-grade REST and gRPC APIs with strict schema validation, rate limiting, caching layers, middleware orchestration, and strongly typed service contracts for scalable internal service-to-service communication.
• Built low-latency data-access layers using PostgreSQL and Redis for hot-path reads and writes.
• Enhanced observability across distributed systems using structured logging, Prometheus metrics, Grafana dashboards, and OpenTelemetry-based distributed tracing.
• Architected event-driven microservice ecosystems using Kafka and Redis Streams.
• Developed real-time fraud-detection and screening services integrated into the bidding path.
• Built and maintained internal bidder CRM and operational platforms using PHP (Symfony 6) and Angular 18+ frontend components, enabling campaign configuration, bidder tuning, operational workflows, and full audit traceability.
• Containerized Node.js services using Docker and deployed them on Kubernetes (AKS), configuring liveness and readiness probes, resource requests and limits, HPA autoscaling, and zero-downtime rolling deployments.
• Improved Kubernetes runtime efficiency through pod-level concurrency tuning, connection-reuse optimization, and graceful-shutdown handling.
• Engineered CI/CD pipelines using Azure DevOps, integrating automated testing, high-throughput load simulations, and progressive delivery strategies.

Master of Science - Computer Science - Middle Tennessee State University (2015 - 2017)

Bachelor of Science - Computer Science - Middle Tennessee State University (2011 - 2015)