Aman Gupta

Lead backend architecture of subscription, entitlement, and metering platforms for multi-tenant, multi-region healthcare products, enabling regulatory compliance and failover resilience with a 99.99% uptime target. Designed and built subscription management and conversation-based usage tracking systems, enabling flexible monetization models, accurate metering, and streamlined customer lifecycle management. Designed and developed an AI-driven skill resolution agent and MCP server that orchestrates retrieval across multiple heterogeneous data stores, improving query accuracy and reducing integration complexity. Architected and delivered a multi-tenant bookmark service adopted across multiple Docquity products, improving feature reuse, cross-product consistency, and platform scalability. Driving infrastructure evolution toward cross-region data replication and high availability, improving system reliability and disaster recovery posture. Collaborate cross-functionally with product, infrastructure, and data teams to deliver backend services aligned with business growth and regional compliance requirements.

Generative AI and Search Platforms Led end-to-end development of a multilingual LLM-powered medical copilot, designed the full RAG architecture including document ingestion pipelines, self-managed PGVector infrastructure with millions of embeddings, and optimized vector retrieval for low-latency, high-accuracy responses, increasing medical content retrieval accuracy by 70%. Designed hybrid retrieval architecture combining semantic embeddings with recency ranking signals, boosting search relevance by 80% based on offline evaluation and user feedback. Integrated LangChain orchestration frameworks to enable contextual reasoning, multi-turn conversations, and domain-specific AI responses at scale. Managed large-scale vector datasets supporting high-speed AI retrieval across millions of medical documents. Distributed Backend and Platform Engineering Designed and scaled microservice-based backend systems in Python, Go, and TypeScript powering communication, AI, and healthcare workflows used by 50,000+ medical professionals. Built high-performance APIs and event-driven services using gRPC, WebSockets, and Kafka, enabling real-time collaboration features and asynchronous processing at scale. Implemented structured logging, distributed tracing, and observability pipelines using Prometheus, Grafana, and OpenTelemetry, reducing production debugging time by 50%. Real-Time Streaming and Messaging Optimized live streaming infrastructure using FFmpeg pipelines, WebSockets, and gRPC, reducing end-to-end latency by 50% and streaming failure rates by 40%. Identified and resolved a concurrency deadlock in an open-source IM server, significantly improving message delivery reliability and system stability for production users. Cloud and DevOps Engineering --- Revamped CI/CD pipelines using Docker, Kubernetes, and Bitbucket Pipelines, reducing build time from 12 minutes to 6 minutes and container image size by 60%, accelerating release cycles. Automated deployment workflows and infrastructure provisioning, improving developer velocity and reducing manual intervention in release processes. Improved deployment stability through Kubernetes rollout strategies, health checks, and structured alerting, reducing incident recovery time significantly. Winner Docquity AI Productivity Challenge for delivering high-impact AI platform features.

Implemented RBAC-based user management system and automated JWT key rotation, improving authentication reliability and security posture across backend services. Built a reusable API pagination framework adopted by multiple teams, improving backend consistency and eliminating duplicate implementations. Delivered performance optimizations across multiple microservices, improving average API response time by 20 to 30%.