Nitesh Timalsina

AI/ML · AWS serverless · Enterprise integration

AI-ready ecommerce and platform work, built to scale without the drama.

I'm Nitesh Timalsina. Eight-plus years in production: chatbots and ML-adjacent flows, breaking monoliths into services, and ecommerce under real load. I default to TypeScript, lean on AWS serverless where it fits, and I'm used to SAP, CMS, and the long tail of vendor APIs.

View resume highlightsContact
Focus
AI + Chatbots
Stack
TypeScript · AWS
Migration
Monolith to microservices
Integrations
Ecommerce · SAP · CMS

A small 3D scene (orbits, glow, a floating core) because big systems still start with a simple idea in motion.

High-impact areas I deliver end-to-end

AI work, ecommerce plumbing, migrations. I care that you can measure the outcome, keep it running, and ship again next week without dread.

AI and machine learning delivery

Chatbots and AI-backed flows where guardrails matter: human review when it should, logging you can actually use, and prompts that don’t sprawl forever.

  • Chatbot architecture
  • Prompt orchestration
  • Operational monitoring

Migration and modernization

Monolith to microservices when the team is ready: clear service lines, contracts people trust, and cutovers that don’t freeze the business.

  • Service decomposition
  • API contracts
  • Progressive cutover strategy

Ecommerce and enterprise integrations

High-traffic commerce wired to SAP, CMS, and search so merchandising and engineering aren’t fighting the same data in two places.

  • SAP connectivity
  • CMS integration
  • Catalog and pricing pipelines

AWS serverless architecture

Serverless-first designs when they fit: APIs, events, data stores, and cost and security called out up front, not as an afterthought.

  • Scalable APIs
  • Event workflows
  • Cloud governance

TypeScript full-stack engineering

Next.js and React apps that stay typed end to end, with shared validation so the UI and backend don’t quietly disagree.

  • Type-safe APIs
  • Reusable UI
  • DX + performance focus

Production operations and reliability

On-call style work: logs, metrics, alerts, and runbooks short enough that someone else could follow them at 2 a.m.

  • Datadog/Splunk
  • Incident response
  • Performance diagnostics