(Yes, I build AI agents that self-correct. My code doesn't always return the favor.)
Full Stack Engineer · AI Systems · Cloud-Native Platforms
📍 Tampa, FL
🎓 M.S. Computer Science @ USF (GPA: 3.95/4.0 · May 2026)
💻 3.7+ Years Production Experience · Optum · BNP Paribas · USF
🔨 Shipping: Enterprise Microservices · Cloud-Native Platforms · Full Stack Apps
🤖 Exploring: Agentic AI · LangGraph Pipelines · RAG Systems · Local LLMs
I'm a Full Stack & AI Engineer who turns:
📦 legacy systems → ⚡ cloud-native platforms → 🤖 self-correcting AI agents
I work across:
- 🏗️ Enterprise full stack — .NET, Java, Spring Boot, React, Angular
- ☁️ Cloud infrastructure — Azure AKS, Cosmos DB, Kafka, AWS
- 🤖 Agentic AI — LangGraph orchestration, RAG pipelines, local LLMs
- 🔐 Security — JWT, OAuth 2.0, RBAC, ASP.NET Identity
Mission: Build systems that don't just respond — they plan, retrieve, verify, and act.
Question → Plan → Retrieve → Verify → Synthesize → Cite
🧑💻 Languages
C# · Java · Python · TypeScript · JavaScript · PowerShell
.NET Ecosystem
- ASP.NET Core · Entity Framework Core · Razor Pages · ASP.NET Identity · xUnit · Moq
☕ Java Ecosystem
- Spring Boot · Spring Security · Spring Data JPA · Hibernate · Maven · Gradle
🐍 Python Ecosystem
- FastAPI · Django · Celery · SQLAlchemy · PyTest · PyPDF2
⚛️ Frontend
- React · Angular · Redux Toolkit · TanStack Query · TypeScript · Tailwind CSS · Vite
🤖 AI & LLM
- LangGraph · OpenAI GPT-4 · Pinecone · LlamaIndex · HuggingFace · Ollama · Llama 3.2
- RAG Pipelines · Prompt Engineering · Tool Calling · PEFT · LoRA · PyTorch
☁️ Cloud & DevOps
- Azure (AKS · Functions · Cosmos DB · Key Vault · Docker · Kubernetes · GitHub Actions · CI/CD App Insights · Service Bus · DevOps)
- AWS (EC2 · S3 · Lambda · ECS · SQS) · Docker · Kubernetes · GitHub Actions · CI/CD
My commits move faster than my coffee cools ☕🐍
To activate the snake, set up this GitHub Action in this repo
🤖 ResearchFlow AI — Self-Correcting Research Agent
- LangGraph multi-node orchestration: Planner → Decomposer → Search → Verifier → Synthesizer
- Web evidence retrieval via Tavily Search API
- Confidence-scored retry logic when evidence quality is low
- Citation-grounded answers — no hallucinated sources
- FastAPI backend · React frontend · Ollama local LLM
🤫 How many nodes in the graph?
Five. Each with a very specific job.
The Verifier's job is to distrust everyone else.
It's the most relatable node.
📄 DocuMind — RAG Platform with Eval Harness
- Semantic search over documents via Pinecone vector DB
- GPT-4 generation grounded only in retrieved context
- LlamaIndex RelevancyEvaluator — actual quality measurement, not vibes
- Sub-second retrieval latency
- FastAPI · React · PyPDF2 · SQLAlchemy
🧪 Did you actually measure quality?
Yes. With an eval harness.
Not just "it seems good."
Revolutionary concept, I know.
🌐 ContextFlow AI — Privacy-First Local LLM Chrome Extension
- Zero cloud API calls — fully local Ollama + Llama 3.2 inference
- Chrome Manifest V3 service worker as persistent agent runtime
- Summarize pages · Ask questions · Explain highlighted text inline
- Floating on-page explanation panel — no tab switching
- TypeScript · React · Tailwind CSS · Vite
🔐 Is this actually private?
Your data never leaves your machine.
Not even to me.
Especially not to me.
🚍 APSRTC Duty Management Portal — Enterprise Workforce Platform
- Spring Boot backend · Spring Security · JWT auth · RBAC across 5 roles
- React + TypeScript frontend · Redux Toolkit · Swagger-documented REST APIs
- Hibernate ORM · PostgreSQL schema design · Docker containerization
- Operational efficiency improved 40%
🚌 Why buses?
Because 40% operational efficiency improvement looks great on a resume.
Also, buses are underrated.
- ASP.NET Core + React platform serving 5,000+ users · 99.6% uptime
- Azure AKS · Redis · Cosmos DB · Application Insights
- 60% faster page load via lazy loading + TanStack Query
- MTTR reduced 35% · 40% of manual workflows automated via PowerShell + Azure Functions
- Kafka pipelines at 3,500+ events/min across healthcare microservices
- Angular + RxJS (Observables, BehaviorSubject, AsyncPipe) frontend
- ASP.NET Core APIs · Application Insights distributed tracing
- MTTR reduced 35% via KPI monitoring and degradation alerting
- Spring Boot + React + AWS financial reconciliation platform
- Reconciliation errors reduced 30% · saved 20 hrs/week
- Amazon SQS async processing · PostgreSQL · Celery task pipelines
- Spring Boot + React workforce management app
- JWT auth · RBAC · Hibernate ORM · Swagger REST APIs
- Operational efficiency improved 40%
🏆 Any awards?
Glad you asked.
Star Performer recognition at Optum.
I put it on my wall. And my resume. And now here too.
No shame.
M.S. Computer Science & Engineering — University of South Florida (GPA: 3.95/4.0 · May 2026)
Cloud Computing · Distributed Systems · Machine Learning · NLP & LLMs · Advanced Databases
B.E. Computer Science & Engineering — Amrita Vishwa Vidyapeetham (GPA: 8.31/10)
🥚 Final Easter Egg (you earned it)
You scrolled all the way down.
Respect.
Here's what you should know:
I ship things. I measure things. I fix things when they break at 3am.
I build AI agents that self-correct — which is more than I can say for most software.
Open to Full Stack SWE and AI Engineer roles — onsite, hybrid, or remote anywhere in the US.
F-1 OPT. Available immediately.
Let's build something. 🚀
"A single-prompt LLM is a calculator. An agent with a plan, tools, and a verifier — that's a system."