enterprise integration × ai agents

Dhanushka Madushan

Technical Team Lead at Enactor, building retail commerce systems that run in stores worldwide. Five years at WSO2 building its integration engine before that. I write and speak about making AI agents work safely with real enterprise systems.

LinkedIn GitHub YouTube DZone Medium X

Focus

Integration architectureAI agents & MCPEvent-driven systemsRetail & POSKafkaCloud nativeEngineering leadership

Book

Cloud Native Applications with Ballerina

Cloud Native Applications with Ballerina Packt, 2021 · 498 pages. Building secure, observable cloud-native services with Ballerina Swan Lake, from language basics to CI/CD. Also on O'Reilly Learning and Amazon.

Writing & talks

Why AI agents break in the enterprise, and what integration middleware already solvedComing soon

WSO2 Synapse HandlersMedium · how to hook into a message flow inside an integration engine

Gentle introduction to Envoy Proxy and load balancingHackernoon

TechCraftZoneYouTube · how things really work, explained simply

Open source

simple-sqlite · 400+ ★An annotated SQLite backend for learning how a database engine works

github-readme · 460+ ★A collection of GitHub README widgets

WSO2 Micro Integrator & Apache Synapse279 merged pull requests, 2018–2023

Experience

2025 – now
Technical Team Lead, Enactor
Retail commerce platform: POS, estate management, self-checkout
2023 – 2025
Senior Software Engineer, Enactor
2018 – 2023
Software Engineer → Senior Software Engineer, WSO2
Micro Integrator, Apache Synapse, API Manager, Choreo
2017 – 2018
Software Engineer, Zeptolytics
2012 – 2016
BSc Eng, Computer Science & Engineering, University of Moratuwa
Two peer-reviewed papers on data-visualization recommendation (ICCIP 2016, IEEE MERCon 2017)

Speaking

Open to talks, podcasts and meetupsTopics: AI agents meet enterprise integration · lessons from 10 years of integration middleware · retail systems at peak scale. Reach me on LinkedIn.