Nikhil Rajput
Full-Stack AI Engineer
AI Automation · GenAI · Backend Engineering
- +91 7049496534
- dev@nik6348.in
- linkedin.com/in/nik6348
- nik6348.in
- Madhya Pradesh, India

Profile
Full-Stack AI Engineer with 2+ years building AI products, automation systems and backend platforms for clients in Australia, South Africa and India. I take a business problem from the first messy conversation through architecture, build, testing, deployment and the production changes that follow — with the deterministic rules, guardrails and human handoffs that keep an AI system trustworthy once real users reach it.
Core skills
Experience
Lead AI automation and full-stack delivery across architecture, LLM workflows, backend services, integrations, testing and production iteration.
- Design and ship AI agents, RAG pipelines, structured extraction, tool calling, guardrails and human-in-the-loop workflows for client systems in production.
- Own the surrounding platform — Python, FastAPI, Node.js, PostgreSQL, Redis, n8n, Playwright and AWS — so the intelligent layer sits on infrastructure that holds.
- Set the reliability practices the team works to: structured outputs, validation, provider fallback, observability and defined escalation paths.
Built backend and full-stack applications for international clients across Australia, South Africa and other markets.
- Delivered REST APIs, webhooks, third-party integrations, dashboards, scraping pipelines and data workflows end to end.
- Worked across Node.js, Express.js, Python, FastAPI, React, Next.js, PostgreSQL and MySQL, including deployment and production support.
Started close to product quality: testing real workflows while contributing to development and debugging.
- Ran functional, API, regression and workflow testing; reproduced defects, validated fixes and pushed edge cases back into the build.
- Built the habit of reading a product as its failure modes — still the first thing I do on a new system.
What I'm relied on for
- Takes ambiguous business requirements to a shipped, supported system
- Draws the line between deterministic rules and model reasoning, deliberately
- Owns frontend, backend, data, agents, infrastructure and deployment
- Designs for failure: validation, fallback, observability, human escalation
- Works directly with international clients across timezones