Validated competency in developing, deploying, and debugging cloud-native applications on AWS. Core focus on serverless architectures (Lambda, API Gateway), containerization (ECS Fargate, ECR), security (IAM, Secrets Manager, KMS), and CI/CD pipelines.

Validated skills in architecting data ingestion pipelines, transformations, storage architectures, and big data orchestration. Expertise across S3 Data Lakes, Glue, Athena, Redshift, Kinesis streaming, and security governance.

Enterprise-grade real-time ride-hailing and fleet operations platform with nationwide Indian coverage.

• Real-Time Navigation: Turn-by-turn road geometry with OSRM & vehicle heading rotation.
• Distributed Engine: Redis Spatial GEO driver tracking + Apache Kafka telemetry stream.
• Security: 20-meter proximity PIN-locking & automated surge pricing with TTL.
• Stack: React 18, Node.js, Socket.IO, Redis, Kafka, MongoDB 2dsphere, Docker.

Production-grade AWS infrastructure and automated CI/CD deployment designed with Infrastructure-as-Code.

• Containerized Runtime: FastAPI service running on AWS ECS Fargate behind an Application Load Balancer.
• Storage & Edge: RDS PostgreSQL in private subnets, S3 presigned uploads, CloudFront CDN.
• Serverless Pipeline: S3 Event-triggered Lambda thumbnail generation.
• Stack: Terraform, AWS (ECS, RDS, S3, Lambda, CloudFront, IAM), Docker, GitHub Actions OIDC.

Comprehensive GenAI architecture toolkit implementing modern LCEL runnables, prompt pipelines, and document loaders.

• LCEL Mastery: RunnableSequence, RunnableParallel, and RunnablePassthrough chain orchestration.
• Deterministic Output: Pydantic schema enforcement and structured JSON output parsers.
• RAG Pipeline: Multi-source document loaders (Directory, Web, Lazy loading).
• Stack: Python, LangChain, LCEL, OpenAI GPT-4o, HuggingFace, Pydantic.

Production RESTful API for high-volume event scheduling, user registration, and capacity tracking.

• Robust Business Logic: Concurrency-safe registration checks and capacity limit enforcement.
• Relational Schema: Normalized PostgreSQL relational database design with indexing.
• Analytics: Real-time event statistics, attendance analytics, and dynamic filtering.
• Stack: Node.js, Express, PostgreSQL, RESTful architecture.

Deepfake and biometric spoofing detection service for biometric security and face authentication.

• Deep Learning Model: Fine-tuned ResNet-18 CNN architecture for real vs. spoof classification.
• Inference API: Low-latency Flask REST microservice for live image classification.
• Evaluation: High confidence scoring and normalized preprocessing pipeline.
• Stack: Python, PyTorch, Torchvision, Flask, Pillow.

Modern collaborative task management single-page application with responsive UI and JWT auth.

• CRUD Lifecycle: Dynamic task states, filtering, tagging, and priority management.
• User Security: Secure JWT-based authentication and protected API endpoints.
• Stack: React (Vite), TailwindCSS, Node.js, Express, MongoDB.