Senior Full Stack Engineer – AI, Backend & Platform Architecture at
Nexus Data Solutions
Current
2022
-
Now
AI & Intelligent Systems
• Designed and implemented a production RAG (Retrieval-Augmented Generation) system for enterprise document intelligence, integrating structured databases, PDFs, and internal tools across 20+ data sources. Reduced average retrieval latency by ~55% through embedding optimization, chunking strategies,and vector index tuning.
• Built LLM orchestration pipelines using LangChain, supporting tool calling, memory management, and multi-step reasoning for AI assistants used in customer support, internal operations, and content workflows.
• Delivered AI chatbot systems with context persistence, role-based knowledge access, and fallback routing, handling ~10K daily interactions while improving support resolution time by ~45%.
• Created a controlled prompt configuration framework that allowed nontechnical teams to adjust AI behavior safely without impacting production stability.
Backend Architecture & Scalability
• Architected cloud-native microservices using Node.js, NestJS, Python, and Go, supporting peak loads of ~50K requests per second with horizontal scaling and graceful degradation under traffic spikes.
• Designed APIs consumed by web, mobile, and internal systems, applying contract-first development, versioning strategies, and backward compatibility to support continuous delivery.
• Implemented background processing and async workflows for AI inference, document ingestion, and analytics pipelines
Data, Observability & Reliability
• Built event-driven pipelines using Apache Kafka, processing hundreds of thousands of events per day for real-time analytics, audit logs, and system telemetry.
• Implemented ML model lifecycle workflows including versioning, staged rollouts, A/B testing, and automated quality monitoring for recommendation and prediction services.
• Optimized PostgreSQL performance using read replicas, query refactoring, indexing strategies, Redis caching, and connection pooling, achieving ~70% improvement in critical query paths.
• Implemented centralized logging, metrics, and distributed tracing with Datadog, significantly reducing mean time to detect and resolve production incidents.
Real-Time & Frontend Systems
• Built real-time communication features using WebSockets, including live chat, typing indicators, user presence, and message persistence, supporting 5,000+ concurrent users.
• Developed complex frontend features with React and TypeScript, including scheduling and booking interfaces with drag-and-drop interactions, conflict resolution, and timezone awareness.
• Created and maintained a shared React component library documented with Storybook, reducing UI inconsistencies and accelerating development across multiple products.
Delivery & Infrastructure
• Containerized applications using Docker and deployed to Kubernetes, enabling consistent environments and zero-downtime deployments.
• Built CI/CD pipelines to automate testing, security checks, and releases across multiple services and environments.
• Participated in on-call rotations, post-incident reviews, and system hardening efforts to improve long-term platform stability.
Special Tech Skills: Node.js, NestJS, Python, Go, React, TypeScript, LangChain, LLM APIs, Apache Kafka, PostgreSQL, Redis, WebSockets, Docker, Kubernetes, AWS, Datadog