Agents, Architecture, and Innovation: A Morning of Agentic AI at Docker x Coders HQ Meetup
Post-GITEX, 27 innovators gathered at Coders HQ for the Agentic AI and Security Meetup. The topic? "Agentic AI and Docker." But it was the questions that revealed the real story—engineers already deploying custom models, architects rethinking infrastructure for multi-LLM systems.
Dubai's tech community came alive post-GITEX 2025 as innovators gathered to explore the future of AI agents and security
Last Saturday morning, something special happened in Dubai's bustling tech scene. The Coders HQ—a government-backed initiative and one of the Emirates' most dynamic software houses—opened its doors for an intimate yet powerful gathering: the Agentic AI and Security Meetup, hosted in collaboration with Docker.
Coders Launchpad
Fresh off the energy of GITEX 2025, approximately 27 builders, researchers, and innovators assembled to dive deep into the rapidly evolving landscape of AI agents, multi-LLM architectures, and the intersection of artificial intelligence with cybersecurity.
Setting the Stage
Saif Rana, CEO of The Launchpad, kicked off the morning with a warm welcome that set the collaborative tone for what would become a highly interactive session. His opening remarks underscored what makes Dubai's tech ecosystem unique: the willingness to experiment, collaborate, and push boundaries.
Agents Are the New Microservices: Unpacking the Multi-LLM Era
Next, I delivered a session on "Agents are the new Microservices: Multi-LLM Era,". The talk explored a paradigm shift happening right now in software architecture—where autonomous AI agents are becoming the building blocks of modern applications, much like microservices revolutionized system design a decade ago.
What made this session particularly memorable wasn't just the content—it was the audience engagement. Questions flew fast and thoughtful, revealing a room full of practitioners actively grappling with real-world AI implementation challenges:
"Tell us more about Agent-to-Agent communication?" – A question that opened up discussions on orchestration patterns, message protocols, and the emerging standards enabling agents to collaborate seamlessly.
"Which model to use and their capabilities?" – This led to a nuanced exploration of the rapidly expanding model landscape, helping attendees navigate the trade-offs between different LLMs for specific use cases.
"How can I add my own MCP server to your MCP Catalog?" – Developers in the room were eager to contribute, demonstrating the open, collaborative spirit that defines the Model Context Protocol community.
"How is DMR different from Ollama?" – A technical deep-dive that resonated with engineers looking to understand local model deployment options.
"I work completely on the Cloud Platform. How is MCP Toolkit beneficial to my day-to-day job?" – This question bridged theory and practice, showing how these tools translate into productivity gains for cloud-native developers.
"What frameworks does Agentic Compose support?" – Attendees were keen to understand integration paths with their existing tech stacks.
"I have my own custom model in HuggingFace. Can I use DMR?" – The sophistication of this question highlighted the maturity level in the room—these weren't just early adopters, but builders already training and deploying custom models.
Docker Captain Takes the Wheel
Esteban Maya, a Docker Captain, brought his expertise to the stage with insights into Docker cagent—A powerful, easy to use, customizable multi-agent runtime that orchestrates AI agents with specialized capabilities and tools, and the interactions between agents. cagent is a powerful framework for building and orchestrating AI agents. Unlike simple chatbots, cagent agents can:
- Take Actions: Use tools to interact with files, run commands, browse the web, and more
- Work in Teams: Coordinate with specialized sub-agents for complex projects
- Remember Context: Maintain persistent memory across conversations
- Think and Plan: Use structured reasoning to solve complex problems
- Integrate Anywhere: Connect to external services via the Model Context Protocol (MCP)
An agent on its own can't do much other than just chat with you. To make agents more useful we need to give the agent tools that it can use to gather information and execute actions. He demonstrated 3 builtin generic agentic tools
thinktodomemory
The think tool gives the model the ability to stop and think. It's a kind of a whiteboard that the model can use to jot down its thoughts. The memory tool gives the agent the ability to remember things about the user. The todo toolset instructs the model to use its todo-tracking tools when it needs to do a complex task. This tool can help the model keep in line while it's working on a complex task.
The Power Behind Coders HQ
The morning culminated with Rama Sbeinaty, Co-Founder of Coders Launchpad, sharing the vision and mission driving this remarkable organization. Her credentials speak volumes: a speaker at GITEX, Microsoft Reactor, and AWS Femme Force, Rama specializes in scalable web services and backend architecture consulting across on-premise and Azure Cloud environments.
But beyond the technical prowess, what shone through was Rama's passion for innovation and community-driven growth. Coders HQ isn't just building software—they're building the ecosystem itself, advising and leading initiatives that shape the future of tech in the UAE. Their work with visionary public and private organizations across the UAE and beyond demonstrates the real-world impact of government-backed innovation done right.
Why This Meetup Mattered
In an era where AI conferences can feel like echo chambers of hype, this meetup cut through the noise. With an intentionally curated group of 27 participants, the event fostered genuine dialogue, technical depth, and practical problem-solving.
The questions asked weren't hypothetical—they came from engineers deploying agents in production, cloud architects rethinking their infrastructure, and researchers pushing the boundaries of what's possible with multi-LLM systems.
Looking Forward
As Dubai continues to position itself as a global AI hub, events like these serve as the connective tissue between vision and execution. The collaboration between Docker and Coders HQ exemplifies what happens when open-source innovation meets government-backed institutional support and private sector agility.
The conversations started last Saturday won't end there. They'll continue in Slack channels, GitHub repositories, and production deployments across the region. They'll influence architectural decisions, model selection, and the next generation of AI-powered applications being built in the UAE and beyond.
When Conversation Becomes Collaboration
Perhaps the most powerful testament to the meetup's impact came after the formal sessions ended. What started as a constructive lunch chat turned into something bigger—building an Agentic AI app for tokenization of tangible Law assets with Jeanina Awni, Director of Technology & Innovation at AlMikial Law Firm (شركة المكيال للمحاماة), and Docker Captain Esteban Maya Cadavid.
This is exactly what Dubai's tech ecosystem enables: the serendipitous collision of domain expertise (legal tech), infrastructure knowledge (Docker/containerization), and cutting-edge AI capabilities. A conversation over lunch transformed into a concrete collaboration that could revolutionize how legal assets are managed and tokenized using agentic AI systems.
It's proof that the best outcomes from tech meetups don't always happen on stage—they happen in the hallways, over coffee, and yes, over lunch, when the right minds come together with a shared vision.
About Coders HQ: A government-backed initiative and one of Dubai's most dynamic software houses, Coders HQ continues to drive real impact across the tech ecosystem in the UAE, working with visionary organizations to turn innovative ideas into production-ready solutions.
Interested in joining future meetups or collaborating with the community? Follow The Launchpad and Coders HQ for upcoming events.
