Neeraj G

About Neeraj G

I am an AI full-stack developer and Machine Learning with more than four years of experience building production-grade, end-to-end AI solutions. I recently completed my Master of Science in Computer Science at the University of Dayton and currently work as a contract AI Software Engineer at a major financial institution, designing and shipping LLM-powered copilots, RAG pipelines, and agentic workflows used by operations and risk teams.My core strength is combining deep Python and cloud engineering with hands-on React and Node.js to deliver secure, scalable AI products. I have designed and implemented agent-based AI assistants, retrieval-augmented generation flows, and workflow automations using OpenAI, Azure OpenAI, Google AI Studio, and Vertex AI, orchestrated with frameworks like LangChain, LangGraph, and LlamaIndex on AWS, Azure, and GCP.On the backend, I build FastAPI and Express microservices, integrate with enterprise authentication and logging, and wire everything into CI/CD pipelines using Git, GitHub Actions, Docker, Kubernetes, and Terraform. On the frontend, I create React-based UIs that make AI decisions explainable, auditable, and easy for business users to trust and control.I enjoy working in lean teams with minimal direction, turning vague ideas into working prototypes, and then hardening them into monitored, reliable production services. I have shipped AI features and workflow automations that reduce manual triage time, cut investigation time per ticket, and improve adoption of AI tools across business teams.I am especially interested in roles where I can:• Build LLM copilots, RAG systems, and agentic workflows end-to-end• Own full-stack implementation (Python, Node.js, React) on cloud platforms• Automate business workflows with robust error handling, monitoring, and observability• Collaborate with product, data, and engineering teams to turn real-world processes into AI-powered solutions

Experience

1. Ai Developer

   Capital One

   Aug 2024 — Present

   Designed and deployed Generative AI and Agentic AI architectures on Azure OpenAI and LangChain, supportingmulti-step LLM agent and sub-agent reasoning pipelines consumed by operations and risk analytics teams.• Architected enterprise-scale RAG pipelines using Pinecone vector databases and OpenAI embeddings with groundingconfigurations, improving response accuracy by 30%+ and reducing hallucinations in production.• Built an AI ticket triage assistant using Azure AI Studio agent orchestration, automating classification and routing andreducing manual investigation effort by 20%+ across operations teams applying AIOps practices for automated incidentdetection and operational efficiency.• Implemented and managed MLOps pipelines to automate model training, deployment, monitoring, and lifecycle management;tuned retry/fallback/timeout thresholds to maintain SLA compliance across concurrent AI workloads.• Developed Knowledge Graph integrations and GraphRAG patterns to enhance multi-hop reasoning over enterprise knowledgebases, improving semantic search accuracy for structured domain data.• Built and maintained Ansible roles to automate provisioning and deployment of Azure AI resources acrossprod/test/experimental environments, reducing manual setup time by 40%; enforced Infrastructure as Code (IaC) practicesvia Terraform.• Engineered token management strategies and cost observability dashboards for GenAI workloads; designed sharedprompt templates and routing abstractions enabling multiple product teams to onboard new tools with minimal boilerplate.• Partnered with data scientists to define offline evaluation datasets, LLM safety guardrails, and model interpretabilitystandards, applying ethical AI practices and hardening prompts prior to production rollout.• Collaborated with architects on environment separation strategies and regional Azure subscription architectures for secure.

Education

• University of Dayton

  Master of Science - MS, Computer Science