Mohsin R.Private profile: Upwork shows its earnings only to signed-in members, so we check it every few weeks, not daily. MRR here is a rough estimate (Upwork's six-month earnings ÷ months active). Last read Oct 25, 2026.

I build production AI agents and AI native SaaS products. I typically own the complete system, from agent orchestration and retrieval through backend APIs, frontend and deployment. My work usually sits where AI engineering meets real product development. I handle the complete system, including agent orchestration, retrieval pipelines, Python and Node.js backend services, APIs, databases, asynchronous processing, frontend integration, third-party services, deployment, and production reliability. I’ve built and shipped systems involving: • AI Agents and multi-agent workflows with LangGraph and LangChain • RAG pipelines for documents, knowledge bases, and internal company data • OpenAI, Claude, Gemini, and other LLM integrations • MCP servers, tool calling, structured outputs, memory, and human approval workflows • Python/FastAPI and Node.js backend services, REST/GraphQL APIs, and microservices • Async processing with Celery, Redis, queues, workers, and event-driven workflows • React and Next.js applications connected to AI and backend services • PostgreSQL, pgvector, Pinecone, Supabase, MongoDB, and Redis • SaaS authentication, subscriptions, RBAC, webhooks, background jobs, and multi-tenant architecture • AWS deployments, Docker, serverless services, monitoring, and production debugging A few examples of the type of work I’ve handled: On an AI SEO platform, I worked on agent-based workflows where specialized agents handled auditing, strategy, and content generation. The architecture included shared state, persisted execution, human approval steps, backend APIs, and recovery mechanisms for long-running workflows. For document AI products, I’ve built Python-based ingestion and RAG pipelines that extract structured information from PDFs and business documents, validate LLM outputs against schemas, store normalized data in PostgreSQL, and expose the resulting intelligence through APIs and product interfaces. On Node.js and TypeScript systems, I’ve built API layers, third-party integrations, webhook processing, authentication flows, real-time services, background jobs, and business logic for production SaaS applications. I’m comfortable working across Express, NestJS, Next.js server-side services, AWS Lambda, PostgreSQL, MongoDB, Redis, and event-driven architectures. I’ve also built AI SaaS products where the LLM is only one part of the system. The harder engineering work is often around reliable tool execution, backend orchestration, permissions, state management, queueing, observability, cost control, and making multi-second AI operations feel responsive in the UI. I use Claude Code extensively in my development workflow for implementation, refactoring, debugging, test generation, repository analysis, and working across larger Python, Node.js, and TypeScript codebases. I’m also comfortable inheriting AI-generated or rapidly developed systems, identifying architectural problems, and hardening them for production without unnecessary rewrites. My primary stack: Python | FastAPI | Django | Node.js | TypeScript | NestJS | LangGraph | LangChain | RAG | OpenAI | Claude | MCP | PostgreSQL | pgvector | MongoDB | Redis | Celery | React | Next.js | AWS | Docker I’m usually a good fit when you need someone who can own more than the prompt or model integration. I can take an AI feature or SaaS product from architecture through Python/Node.js backend development, frontend integration, deployment, and production support while keeping the system maintainable as it grows. If you’re building an AI agent, RAG platform, AI-enabled SaaS product, or a production backend in Python or Node.js, feel free to share the architecture or current codebase and I can quickly identify the right starting point.