Amul Sharma

I’m a Full Stack Developer with around 2 years of experience building scalable web applications, with a strong focus on backend development using Node.js, Express, and REST APIs. I like owning features end-to-end and turning third-party APIs and webhooks into reliable, production-ready functionality.

At Real Vibe, I built an AI-driven Google Ads reporting layer using the Model Context Protocol (MCP), connecting an LLM directly to the Google Ads API with OAuth 2.0 and manager-account routing so the team can pull live metrics across multiple accounts in natural language. I also co-developed an automated sales agent that processes real-time messages via the Meta WhatsApp Business API webhooks, migrated a legacy Angular app to React for a more maintainable codebase, and handled AWS deployments on EC2 and RDS to keep production running reliably.

My project work blends modern web engineering with applied ML. In BigLead, a multi-tenant SaaS CRM, I architected Postgres row-level security isolation, built Meta lead capture with signature-verified webhooks and idempotent ingestion, and integrated Razorpay subscriptions and transactional email. Separately, I developed a pneumonia detection model using DenseNet-121 and TensorFlow, reaching 96.68% test accuracy—proof that I enjoy pairing strong engineering practices with data-driven outcomes.