All authors
Sujith PS

Co-founder & CTO

Sujith PS

Co-founder and CTO of Kiebot. Specialises in reactive systems, distributed architecture, and the Agile delivery practices that keep complex platforms shipping every week.

Kerala, India18+ years experience13 articles

About Sujith

Sujith PS is the Co-founder and Chief Technology Officer at Kiebot. He has spent close to two decades architecting backends for high-traffic platforms, building reactive event-driven systems, and coaching engineering teams on Agile delivery. He has owned the technical strategy for products in healthcare, fintech, e-commerce, and SaaS, with deep hands-on time in Node.js, NestJS, Spring, and Kubernetes. He writes about reactive programming, clean code, software architecture, and the lessons learned from running an engineering org that ships to production multiple times a day.

Recent highlights

  • 1

    Architected the reactive platform behind Peditoh, Kiebot’s healthcare product.

  • 2

    Designed multi-tenant SaaS backends used by retailers across the GCC.

  • 3

    Built Kiebot’s internal engineering playbook for code review, on-call, and release engineering.

Articles by Sujith

Node.js Development to the Next Level
4 April 2024

Node.js Development to the Next Level

Node.js has revolutionized server-side JavaScript development with its efficiency and scalability.

Read article →
Mastering Agile Testing
26 March 2024

Mastering Agile Testing

Key principles and practices for high-quality software delivery in an Agile environment.

Read article →
Reactive Programming: A Deep Dive into Modern Architecture
20 January 2026

Reactive Programming: A Deep Dive into Modern Architecture

Move beyond imperative code. Learn how to build resilient, elastic, and message-driven systems that scale effortlessly.

Read article →
The Art of writing clean code
11 November 2020

The Art of writing clean code

Readable code is easier to write, understand, and maintain. Learn the principles of clean engineering.

Read article →
Thinking in AI: From Deterministic to Probabilistic Systems
05 February 2026

Thinking in AI: From Deterministic to Probabilistic Systems

Software engineering is changing. We are moving from strict logic rules to probability and context. How do you architect for uncertainty?

Read article →
Designing AI Applications That Survive Production
25 March 2026

Designing AI Applications That Survive Production

A working demo is easy. A production AI application that is reliable, safe, and affordable is hard. Here is the architecture I keep returning to.

Read article →
RAG in Production: A Practical Guide for 2026
02 April 2026

RAG in Production: A Practical Guide for 2026

Retrieval-Augmented Generation sounds simple in a demo and gets ugly in production. Here is the version that actually works.

Read article →
The LLM Stack in 2026
05 April 2026

The LLM Stack in 2026

Six layers that make up a serious LLM application. Skip one and you will feel it on day 30.

Read article →
LangGraph for Stateful Agents
11 April 2026

LangGraph for Stateful Agents

Agents that take more than one step need state, branches, and checkpoints. LangGraph is the cleanest way we have found to express that.

Read article →
HermisAgent and the Open Agent Platform
18 April 2026

HermisAgent and the Open Agent Platform

HermisAgent is the runtime that powers Open Agent, our platform for building autonomous AI workers. Here is the architecture and the trade-offs.

Read article →
Open-Source LLMs That Actually Matter in 2026
22 April 2026

Open-Source LLMs That Actually Matter in 2026

Open weights are no longer a hobby. Here is the shortlist we recommend and the use cases each model wins.

Read article →
Claude Code in 2026: What Is Actually New
29 April 2026

Claude Code in 2026: What Is Actually New

Claude Code has shipped fast in the last year. Here are the changes that matter on a real project, not on a launch slide.

Read article →
Gemini CLI vs Claude Code: An Engineer's Comparison
02 May 2026

Gemini CLI vs Claude Code: An Engineer's Comparison

Two great AI coding agents, two very different philosophies. Here is how they compare on the work we actually do.

Read article →

Questions Sujith gets asked

When does Sujith recommend reactive programming over a traditional REST stack?

Reactive shines when you have high-fan-out workloads, websocket-heavy products, or systems that must hold steady under unpredictable traffic spikes. For a CRUD admin tool, classic REST is still the right tool.

How does Kiebot’s engineering org keep code quality high while moving fast?

Three non-negotiables: every PR is reviewed by a senior, CI runs the same checks that block production, and we treat tests as part of the deliverable, not an afterthought.