Full-stack engineer with a product mindset, building AI-native workflows that solve real team problems.
I work across frontend, backend, and orchestration layers to ship practical AI systems that people actually use — and occasionally over-optimize workflows for fun.
- Find the highest-friction workflow in the product/user journey
- Prototype quickly with full-stack + AI tooling
- Add reliability layers (fallbacks, validation, observability)
- Ship, measure, and iterate based on actual usage
I care less about hype and more about: Did this save time, improve quality, or unlock a new capability for users?
Interface: React, Next.js, TypeScript, Tailwind
Backend: Node.js, APIs, service integration, workflow logic
AI Layer: LLM integrations, prompt/system design, agent orchestration patterns
Product Ops: experimentation, iteration loops, outcome-oriented delivery
I prefer capability-based systems over stack flexing — the goal is reliable, useful product behavior.
- Execution > theory: ship small, validate fast
- Reliability first: graceful failure beats brittle magic
- Outcome-driven AI: every AI feature should map to a business/user win
- Keep it practical: if a workflow can’t be maintained, it’s not production-ready
I’m moving deeper into:
- agentic product architecture
- system design learning for real-world product workflows
- AI-assisted engineering systems that improve team velocity
If you’re building in this space, I’m always open to exchanging ideas and implementation patterns.
- LinkedIn: linkedin.com/in/sukrit-saha-b6117a242
- X / Twitter: x.com/SukritSaha11
- Portfolio / Website: sukritsaha.in
- Email: [email protected]