NDR – The AI Conf 2021

When teams adopt Microservices with an understanding of the structure of the architecture, but not of how to get all the pieces to communicate, it is all too easy to accidentally create a distributed Big Ball of Mud. Neal introduces a new measure, the architecture quantum, to help analyze and identify communication boundaries and define static and dynamic coupling. Then, the session provides tools – integrators and disintegrators – to help architects iterate towards the correct granularity for their Microservices for static coupling. Next, for dynamic coupling, architects must understand when to choose synchronous versus asynchronous communication between services, consistency, and coordination to ultimately analyze transactional sagas; this talk describes eight possible sagas and when each is applicable.

A common saying by software architects is “that’s an implementation detail”. All too often we treat software architecture and implementation as two separate things, where implementation is something that happens once a software architecture is defined. In fact, it’s the other way around: software architecture should be viewed as a first draft, where implementation reveals more details and refinements. In this session Mark Richards discusses the intersection of architecture and implementation and how the two must be in constant alignment to achieve success, demonstrating along the way why architecture is a critical element of any system. Through real-world examples, he shows how implementation can easily get out of alignment with the architecture, causing the system to fail to achieve its desired goals. He then shows some techniques and tools to help ensure the alignment between architecture and implementation.

Production agentic AI needs more than agents.
It needs structure. It needs boring stuff: observability, modularity, data profiling, and monitoring.
Otherwise, it becomes chaos with a personality, fancy demos that fail silently in production.
In this talk, we’ll show you how old-school software and MLOps principles are the secret weapon for building real, scalable, and reliable agentic systems.
No hype. No buzzword bingo.
Just field-tested thinking and hard-learned lessons from production AI deployments.
What you’ll take home:
* A framework to cut through the noise and think clearly about agentic architecture
* How to debug, observe, and monitor agents like real software systems
* How to avoid the trap of shiny tools and focus on system design that actually works
* A checklist for building agentic AI that doesn’t crash after your first user touches it

Leadership Coach and author Andrei Postolache talks about the attitudes, behaviours and skills that high performance Individual Contributors and Leaders need to succeed in today’s world. Based on his work with hundreds of teams and individuals, he narrows down the essential organizational, communicational and inter-relational skills that truly make the difference.

Coding has always been more than just writing lines of code; it’s about solving puzzles, creating solutions, and adapting to challenges. But what happens when AI begins to tackle those puzzles as well? How will your role evolve in this new landscape?

As AI integrates deeper into our toolsets and workflows, the real revolution isn’t about simply learning new technologies. It’s about fundamentally changing the way we think, how we architect solutions, and our entire approach to software development. This talk cuts through the AI hype and zeroes in on your growth as a developer.

Discover practical strategies to leverage AI beyond mere automation, unlock untapped creativity, strategic thinking, and problem-solving. It’s time to not just write code, but to reforge it, leveraging AI as a powerful ally in your journey towards mastery.

…