AI Engineer | GenAI Specialist

Architecting the Future with Generative AI

Building Production-Grade GenAI Solutions. Transforming innovative ideas into scalable AI applications with LLMs, RAG systems, and cloud-native architectures. From concept to deployment—making AI work for real-world impact.

View Projects Get in Touch

GenAI Stack & Cloud Technologies

Building Production‑Ready GenAI Systems

Focused execution across frameworks, scalable backends, cloud deployment and human‑centered interfaces.

GenAI & Agentic Frameworks

Designing intelligent, retrieval-augmented & tool-using systems

LangChain / LangGraph (agent orchestration)
LlamaIndex (structured + hybrid RAG)
Hugging Face & Transformers
TensorFlow • PyTorch (DL training)
LangSmith / LangFuse (eval & tracing)
AutoGen (multi-agent collaboration)

RAGHybrid RetrievalPrompt EngineeringEvaluationFine-tuningTooling

Backend & Core Engineering

High-performance APIs & services powering AI workloads

Python (async, packaging, perf)
FastAPI (low-latency inference services)
C++ (performance-critical components)
SQL + Query Optimization
Vector Search Pipelines (FAISS, Pinecone, Chroma)
Neo4j (graph reasoning) & MongoDB

API DesignStreamingBatch PipelinesCachingObservabilityTesting

Cloud, DevOps & MLOps

Reliable, observable & cost-aware production deployment

AWS (Bedrock, Sagemaker, ECS, Fargate, Lambda, S3)
GCP Vertex AI (managed LLM + pipelines)
Containerization & Orchestration (Docker, Fargate)
CI/CD Automation (GitHub Actions)
Model & Data Lifecycle (versioning, eval, rollout)
Dataiku & Snowflake integration

ScalingCost OptimizationMonitoringTracingDrift DetectionSecurity

Frontend & Experience Layer

Human-centered interaction for AI capabilities

React & Next.js (SSR/ISR)
Streamlit (rapid experimentation UIs)
Real-time UX (websockets, streaming tokens)
Design Systems & Accessibility
State Management & Data Fetching
Visualization & Insight Surfaces

Prompt BuildersChat InterfacesAnalytics DashboardsPlaygrounds

Architecting trustworthy AI: orchestrating agents, hybrid retrieval, hardened inference APIs & reproducible deployments that balance latency, accuracy and cost.

Featured Projects

Production-grade GenAI solutions deployed across healthcare, finance, insurance, and enterprise domains

LangChainAWSRAGPython

GenAI-In-A-Box Framework

Proprietary framework that accelerated GenAI application delivery by 50%+. Engineered plug-and-play pipelines for Modular, Hybrid, and Advanced RAG workflows using Python, LangChain, and LlamaIndex. Deployed fault-tolerant services on AWS ECS Fargate with persistent VectorDB storage ensuring 99.9% uptime.

View Project

GeminiRAGFastAPIStreamlit

Insurance Agent (Hybrid RAG)

Domain-aware AI assistant powered by Gemini Pro that improved policy information retrieval accuracy by 30%. Implemented hybrid RAG architecture combining dense and sparse retrieval for optimal context understanding and response generation.

View Project

LangChainMongoDBAgentFastAPI

Hospitality Agent (RAG + Function Calling)

LangChain Agent with full CRUD operations on MongoDB backend, automating 60% of manual booking tasks. Integrated function calling capabilities for real-time data manipulation and intelligent booking management.

View Project

NLPRAGPythonML

HR Candidate Pre-Screening RAG

Rerank-enhanced RAG system to extract structured insights from resumes, reducing initial screening time by 40%. Leveraged advanced NLP techniques for semantic understanding and candidate matching.

View Project

RAGChatbotAWSReact

Website Concierge Chatbot

Advanced RAG assistant deployed on company website, increasing user engagement by 25% and improving lead qualification. Features multi-turn conversation handling and context-aware responses.

View Project

Deep LearningHealthcarePythonML

Chronic Kidney Disease Prediction

Hybrid deep learning model combining ANN with Cuckoo Search optimization achieving 92% prediction accuracy. Designed data preprocessing pipeline reducing noise by 40% through systematic hyperparameter tuning.

View Project