I build AI systems end to end, from agent architectures and infrastructure to the interfaces people actually use.
I build AI systems across the stack, from agent architectures and LLM workflows to infrastructure, evaluation, and production interfaces. I’m interested in the engineering challenges that emerge when AI moves beyond prototypes and into real-world use.
My work focuses on making AI systems reliable, observable, and practical to operate. That means thinking beyond the model itself and working across system design, deployment, testing, and the interfaces that connect systems with people.
I also contribute beyond the code through open source, technical writing, talks, workshops, and mentorship. I enjoy sharing what I learn and helping other engineers build better AI systems.
Building agentic systems, orchestrating LLM workflows, and designing evaluation pipelines for production applications.
Developing tools, contributing to projects, and sharing technical implementations across AI and software engineering.
Speaking at conferences and meetups, leading technical workshops, and sharing lessons from building AI systems.
Writing about agentic systems, engineering practices, and lessons from building AI in production.
Creating technical tutorials and walkthroughs on AI systems, software engineering, and cloud infrastructure.
Organising technical meetups, mentoring engineers, and creating spaces for people and AI agents to learn and build.
Roles spanning AI engineering, software development, and building production systems.
Developing and deploying AI systems for production applications. Working across agentic systems, LLM workflows, backend services, and cloud infrastructure to integrate AI into existing products and build new solutions.
Built software applications and developed proofs of concepts (PoCs) for AI-driven features. Gained hands-on experience across backend development, cloud technologies, and the early stages of building AI systems.
Worked on mobile app development projects using Flutter and Dart. Collaborated with designers and developers to create user-friendly applications. Acquired experience in mobile app lifecycle and deployment.
Developed custom software solutions for various clients. Focused on creating user-friendly applications and optimizing existing systems. Developed expertise in project management and client communication.
The tools behind the systems, experiments, and products I build.
Projects I build, maintain, and contribute to.
One agent core. Swap the model, swap the surface, swap the host.
Where I studied and what I learned along the way.
Sharing ideas, bringing developers together, and contributing to the tech community.
Communities I organise with, contribute to, or volunteer for.
Contribute in organizing meetups, events, and bootcamps to connect AI enthusiasts and professionals in Sri Lanka.
Assisted in organizing events and meetups to foster collaboration and knowledge sharing among developers in Sri Lanka.
Engaged in community building and knowledge sharing through workshops, webinars, and tech talks.
Participated in community events, workshops, and meetups to promote learning and collaboration among students and professionals.
Talks and workshops I've delivered, with the most recent first.
A practical workshop focused on deploying agentic AI applications using modern cloud architecture. Participants learned how to build reliable background agent workflows, connect external tools and databases, and deploy a working AI application.
A hands-on workshop introducing the Microsoft Agent Framework for building AI-powered agents and intelligent workflows. Participants learned how to create agents, integrate tools and MCP, manage conversations and memory, and build practical AI applications.
An overview session introducing Microsoft Foundry as a unified AI development platform. Participants explored how to build, manage, and securely deploy enterprise-grade AI solutions.
A project-based workshop applying Azure AI Foundry to real AI agent scenarios. Participants gained practical experience in building, testing, and scaling intelligent agents.
Notes I write on AI systems, engineering, and production.
A technical breakdown of the Model Context Protocol (MCP), covering client-host-server topology, stdio vs. SSE transports, schema negotiation, and tool execution lifecycles.
How to prevent context window saturation and token starvation in autonomous AI agents through sliding windows, semantic distillation, and state-delta compaction.
A practical guide to Microsoft's current AI certification portfolio in 2026, covering Azure AI, AI agents, MLOps, Copilot, multi-agent systems, Power Platform, Dynamics, and GitHub Copilot.
How to test non-deterministic AI agents in CI using deterministic assertions, trajectory checks, and structured LLM-as-a-judge rubrics.
Walkthroughs I create on AI systems and software engineering.
This video provides a step-by-step guide to creating a Microsoft Azure free account and accessing $200 in free credits. It explains the account registration process and introduces how beginners can use the available credits to explore more than 55 Azure services, experiment with cloud technologies, and start building real-world projects without an initial cost.
This video provides a step-by-step guide for students to create an Azure for Students account using a voucher and access $100 in Azure credits. It explains the account setup process and helps students get started with Microsoft Azure so they can explore cloud services, build projects, and gain practical experience without an initial payment.
This video provides a practical introduction to working with Azure OpenAI and using chat completion capabilities. It walks through the fundamentals of interacting with AI models, sending prompts, and generating responses, giving developers a starting point for building applications powered by generative AI and conversational experiences.
This video introduces Azure OpenAI and explains how developers can leverage OpenAI models through Microsoft Azure. It provides an overview of Azure OpenAI capabilities, its integration with Azure services, and how developers can use generative AI models to build intelligent applications and experiences.
I enjoy connecting with people who are building, learning, and exploring new ideas. Whether you have a question, an idea, or just want to say hello, feel free to reach out.
[email protected]Social profiles and everything else: links.dileepa.dev