GIDS 2027 Sessions | Software Architecture, AI & Engineering

Java Records remove boilerplate code and make it easier to represent data, but creating and using them comes with nuances and idiosyncrasies that developers nee...

Pause, Approve, Resume: Durable Human Control for AI Agents

Tue, April 27

AI-Native SoftwareArchitecture & Distributed SystemsReliability, Observability & Security

Daniel Oh

Multi-agent workflows become considerably harder when an agent must stop and wait for a human decision. Approval may arrive minutes or hours later, after a rest...

Lessons from Economics for the Software Engineering and Agent Workforce

Tue, April 27

AI-Assisted Software Engineering

Daniel Nadasi

AI is reshaping software engineering, creating uncertainty about how engineering roles will change as increasingly capable systems take on work previously perfo...

After the AI Hype - What’s Real, and What’s Next

Tue, April 27

AI-Assisted Software EngineeringData & Intelligent Systems

Richard Campbell

Artificial intelligence dominates the headlines, and we are in the middle of a major hype cycle. Beneath the attention and investment, however, genuine innovati...

AI-Assisted Legacy Modernization

Tue, April 27

AI-Assisted Software EngineeringBackend Engineering & RuntimesArchitecture & Distributed Systems

Every organisation has them: the Java 8 services nobody wants to touch, the .NET Framework applications quietly holding up a business process, and the dependenc...

Your API Is Not Ready for AI Yet

Tue, April 27

Backend Engineering & RuntimesArchitecture & Distributed SystemsAI-Native Software

APIs designed for human developers often rely on documentation, informal conventions, and manual recovery. Autonomous agents are less forgiving. They can retry ...

Benefits and Pragmatics of Agent-Based Architectures

Tue, April 27

AI-Native SoftwareArchitecture & Distributed Systems

Agents seem to be emerging everywhere. With the capabilities of AI, it has become very easy to create them. The familiar question, “not if we can, but if we sho...

Definition Of Insanity: Timeouts, Retries and Idempotency

Tue, April 27

Reliability, Observability & SecurityArchitecture & Distributed Systems

Sam Newman

“The definition of insanity is doing the same thing over and over again.” This quote, often attributed to Einstein, warns against repeatedly trying something th...

AI Beyond the Keyboard: Speech Synthesis, Recognition, and Accessible AI

Tue, April 27

Front-End & Experience EngineeringAI-Native Software

For many developers, AI has become synonymous with text: prompts, tokens, source code, and chatbot responses. But text represents only a fraction of human commu...

Everything in Tech Is Changing. Are We?

Tue, April 27

AI-Assisted Software EngineeringArchitecture & Distributed Systems

The ground under software is moving quickly. Assumptions on which we built entire practices are changing: that a human is in the loop, that a human consumes the...

Designing Agentic Architecture for AI-Native Systems

Tue, April 27

Architecture & Distributed SystemsAI-Native Software

Nicole Koenigstein

Most software architectures assume that system behaviour is largely determined before deployment. Agentic systems challenge that assumption. When reasoning, too...

Creating High Performance Systems: Lessons From the Trenches

Tue, April 27

Backend Engineering & RuntimesReliability, Observability & SecurityArchitecture & Distributed Systems

Not every application needs high performance everywhere. But when performance is critical, a brute-force approach to increasing speed often does not solve the u...

Your Prompt Is Not a Sandbox: Engineering Lessons from the 2026 Agent Breaches

Tue, April 27

Reliability, Observability & Security

In 2026, AI agents running inside supposedly isolated security evaluations reached the production systems of real organisations. The headlines said “AI escapes....

The AI Control Plane: Governing Agents, Tools, Evidence, Cost, and Human Approval at Scale

Tue, April 27

AI-Native SoftwarePlatform Engineering & DevOpsReliability, Observability & Security

As organisations deploy more agents, individual guardrails become difficult to govern consistently. Different teams may define tool permissions, evidence rules,...

Everyone is a Builder

Wed, April 28

AI-Assisted Software Engineering

Tejas Kumar

Software has long been gatekept by syntax. Today, those barriers are disappearing, opening software creation to a much wider group of people. What does that mea...

Specification-Driven Development by Example

Wed, April 28

AI-Assisted Software EngineeringArchitecture & Distributed Systems

You cannot prompt your way to production. Software development is not about producing code at breakneck speed, and speed without discipline can create more prob...

Designing Codebases for Coding Agents

Wed, April 28

AI-Assisted Software EngineeringArchitecture & Distributed SystemsPlatform Engineering & DevOps

Ragunath Jawahar

Modern tooling makes it increasingly easy to run multiple coding agents against the same codebase, but most codebases were not structured for that way of workin...

Our Data Model Sucks. It's Not 1970 Anymore.

Wed, April 28

Data & Intelligent SystemsAI-Native Software

In 1978, data modeller Bill Kent observed that what we call data does not, and cannot, fully describe reality. It captures the state our software needs to remem...

Fast, Lean Java on Kubernetes: What Actually Moves the Needle

Wed, April 28

Backend Engineering & RuntimesPlatform Engineering & DevOps

Daniel Oh

Java applications running on Kubernetes are expected to start quickly, consume less memory, use compact container images, and scale efficiently. Achieving all f...

Principles for Designing User Interfaces

Wed, April 28

Front-End & Experience Engineering

Think about the last time you used something that worked seamlessly. It may be easier to remember the opposite: a tool, product, application, or appliance that ...

Two Writers, One Model: Safe Shared State for Human-Agent Interfaces

Wed, April 28

AI-Native SoftwareFront-End & Experience EngineeringArchitecture & Distributed Systems

Martin O'Hanlon

What happens when a person and an AI agent can both change the same live artifact? An agent may report that it has made an edit when no tool call actually ran. ...

Above the Cloud: Building Data Centers in Space

Wed, April 28

Platform Engineering & DevOpsArchitecture & Distributed SystemsReliability, Observability & Security

Richard Campbell

The demand for AI data centers is growing rapidly, placing increasing pressure on power infrastructure. Could putting those data centers in space make sense? Th...

Parameters, Context, and Quantization, Oh My! The Nuts and Bolts of Running, Configuring, and Tuning Local AI

Wed, April 28

Backend Engineering & RuntimesAI-Native SoftwarePlatform Engineering & DevOps

Running open-weights AI models locally can seem straightforward until you encounter unfamiliar technologies and acronyms such as GGUF, MLX, and CUDA. Why can a ...

The Lost Art of Software Design

Wed, April 28

Architecture & Distributed SystemsAI-Assisted Software Engineering

Simon Brown

“Big design up front is dumb. Doing no design up front is even dumber.” This captures the journey from big design up front in the 20th century to emergent desig...

How to Build AI You Can Trust

Wed, April 28

AI-Native SoftwareReliability, Observability & SecurityArchitecture & Distributed Systems

Tejas Kumar

Trust is a fundamental challenge when building AI systems, both ahead of time and while they are running. In this live-coding session, you will explore both sid...

CI/CD in the Age of Continuous AI: What to Automate, Gate, and Contain

Wed, April 28

Platform Engineering & DevOpsAI-Assisted Software EngineeringReliability, Observability & Security

As teams adopt AI for code generation, testing, security review, documentation, and release decisions, the CI/CD pipeline becomes a natural home for continuous ...

Twelve Essential Engineering Practices for AI-Assisted Development

Wed, April 28

AI-Assisted Software Engineering

Using AI to write code is like driving a car: learning to drive defensively and sensibly can help you reach your destination safely and quickly. While much of t...

6 Event-Driven Architecture Patterns You Should Know About

Wed, April 28

Architecture & Distributed SystemsReliability, Observability & SecurityBackend Engineering & Runtimes

Event-driven architecture can provide high levels of responsiveness, scalability, elasticity, and fault tolerance, but it also introduces complexity that can ma...

When Agents Meet Microservices: Surviving Retries and Failure

Wed, April 28

Architecture & Distributed SystemsReliability, Observability & SecurityAI-Native Software

Microservices are commonly designed around bounded requests from predictable clients. Agents change that assumption. They can plan across services, retry after ...

Progressive Collapse: How To Avoid Cascading Failure

Wed, April 28

Reliability, Observability & SecurityArchitecture & Distributed Systems

Sam Newman

Small things going wrong can quickly snowball. A problem that seems minor in isolation can trigger a chain reaction of increasing failures, potentially leading ...

Using AI to Upgrade Applications from One Language to Another: Strategies, Dos, and Don'ts

Wed, April 28

AI-Assisted Software EngineeringBackend Engineering & Runtimes

You have been asked to modernise a high-profile, difficult-to-maintain application by moving it from an older language to one better suited to current needs, an...

Don't Build an Ontology, Harvest It. The Semantic Blueprint

Wed, April 28

Data & Intelligent SystemsArchitecture & Distributed SystemsAI-Native Software

Neal Ford , Mark Richards

The meaning needed for an enterprise semantic layer may already exist across the systems an organisation runs: in its code, schemas, ETL, and the domain knowled...

What Breaks When Research Agents Go to Production?

Wed, April 28

AI-Native SoftwareData & Intelligent SystemsReliability, Observability & Security

Sarang Kulkarni

The hardest part of building production research agents is not generating impressive answers. It is controlling the environment in which they reason. Systems th...

Why does Yesterday's Best Practice Become Tomorrow's Anti-pattern?

Thu, April 29

Architecture & Distributed SystemsAI-Assisted Software Engineering

Software development exhibits a curious pattern: yesterday’s best practice can become tomorrow’s anti-pattern. EJB and SOA were once regarded as best practices,...

A Practical Guide to Creating and Orchestrating Agents

Thu, April 29

AI-Native SoftwareArchitecture & Distributed SystemsBackend Engineering & Runtimes

Building applications with agents raises practical questions. How do you create agents and communicate with them? How do you implement workflows that span multi...

When Models Argue with Each Other

Thu, April 29

Data & Intelligent SystemsAI-Native SoftwareReliability, Observability & Security

Gireesh Punathil

Adding more models might seem like a straightforward way to improve prediction quality, but in practice the results can be very different. While building a high...

Scaling Developer Infrastructure for an Agentic World

Thu, April 29

Platform Engineering & DevOpsAI-Assisted Software EngineeringReliability, Observability & Security

Daniel Nadasi

Google’s source, build, test, continuous integration, and release systems are under unprecedented pressure as software development scales from tens of thousands...

Architecture AntiPatterns and Pitfalls

Thu, April 29

Architecture & Distributed SystemsBackend Engineering & Runtimes

Tuhin Sharma , Soham Dutta

When do you need an architecture? How should you iterate on it as a system evolves? How do you create a microservices architecture, and how do you know whether ...

Purpose-Built Harnesses for Software Factories

Thu, April 29

AI-Assisted Software EngineeringPlatform Engineering & DevOpsReliability, Observability & Security

Ragunath Jawahar

As organisations explore software factories and the use of AI to industrialise software delivery, coding agents introduce a fundamental challenge: unbounded int...

Production-Ready Agents: A Live Architecture Workshop

Thu, April 29

AI-Native SoftwarePlatform Engineering & DevOpsReliability, Observability & Security

Most AI agent tutorials focus on capabilities, but production systems often encounter a different set of problems. Agents that perform well in development can b...

Cobots, Dark Modules and Architecting For AI

Thu, April 29

AI-Assisted Software EngineeringArchitecture & Distributed Systems

Sam Newman

The use of AI in software delivery has been transformative, or more precisely, we are still in the middle of that transformation. It is creating challenges for ...

Modern Enterprise Architecture Practices

Thu, April 29

Architecture & Distributed Systems

Neal Ford

Enterprise architecture is increasingly important in the age of agentic AI. In this session, you will explore how enterprise architecture has evolved, including...

When Healthy Agents Fail: 12 Patterns from Production

Thu, April 29

Reliability, Observability & SecurityAI-Native Software

Tuhin Sharma , Soham Dutta

Production AI agent systems can appear healthy by traditional operational measures while delivering a steadily degrading user experience. Latency remains low, t...

Asynchronous Programming in Spring: Past to Present

Thu, April 29

Backend Engineering & RuntimesArchitecture & Distributed Systems

Building scalable Spring applications often depends on choosing the right approach to asynchronous programming. In the past, developers relied heavily on reacti...

The Undersea Infrastructure

Thu, April 29

Architecture & Distributed SystemsReliability, Observability & SecurityPlatform Engineering & DevOps

Richard Campbell

Bandwidth is the oxygen that feeds the internet, and the undersea cable network provides that bandwidth across the world’s oceans. These cables span thousands o...

Small Models, Huge Impact: The AI in Your Pocket

Thu, April 29

AI-Native SoftwareFront-End & Experience EngineeringArchitecture & Distributed Systems

Cloud AI services have attracted tens of millions of paid subscribers, but the potential reach of AI-ready smartphones is considerably larger. Industry estimate...

Policy Must Commit: Securing Agent Actions Beyond Guardrails

Thu, April 29

Reliability, Observability & SecurityAI-Native SoftwareArchitecture & Distributed Systems

An AI agent can be authenticated, permitted to call a tool, and still take an action the organisation cannot defend. Retrieved instructions may be malicious, ev...

Who Is This Agent Acting For? Authorization for MCP and Agent Tools

Thu, April 29

Reliability, Observability & SecurityAI-Native SoftwareBackend Engineering & Runtimes

When an AI agent calls a tool, someone’s authority is being used. But whose? The user who asked, the agent itself, or the platform running it? Get that wrong, a...

Testing in Production: No Kidding

Thu, April 29

Reliability, Observability & SecurityPlatform Engineering & DevOps

“Testing in production” is often used as a tongue-in-cheek phrase, but when approached proactively and with the right intent, it can be a powerful way to build ...

Where Does Knowledge Live? Semantic Architecture in Production

Thu, April 29

Architecture & Distributed SystemsData & Intelligent SystemsReliability, Observability & Security

Semantic layers and knowledge graphs promise new ways for AI systems to work with organisational knowledge, but moving from the idea to a production architectur...

Managing Codebases for Machine-Speed Development

Thu, April 29

AI-Assisted Software EngineeringArchitecture & Distributed SystemsPlatform Engineering & DevOps

Daniel Nadasi

As software development scales from tens of thousands of human developers to millions of machine-speed developers, the assumptions behind traditional codebase m...

The Subtle Ikigai of Local AI

Fri, April 30

Architecture & Distributed SystemsAI-Native SoftwarePlatform Engineering & DevOps

The current AI narrative often suggests that bigger is better: trillion-parameter models running in remote data centres, consuming enormous amounts of power, an...

You Have to Understand Architecture to Teach Agents About It

Fri, April 30

AI-Assisted Software EngineeringArchitecture & Distributed Systems

The industry is moving forward with agentic code generation, with both successes and failures along the way. Many of those failures stem from agents lacking kno...

Software Architecture Trade-Off Analysis

Fri, April 30

Architecture & Distributed SystemsBackend Engineering & Runtimes

Neal Ford

There are no universal best practices in software architecture. Every design decision involves trade-offs, which is why the answer to so many architecture quest...

Skill Up to Train Your AI Dragon

Fri, April 30

AI-Assisted Software Engineering

AI is powerful, but its non-deterministic nature means the important question is not whether to use it, but how to use it effectively for application developmen...

Patterns for Deterministic AI in the Enterprise

Fri, April 30

AI-Native SoftwareReliability, Observability & Security

AI has significant potential in the enterprise, but reliability remains a challenge. Ask an LLM for last quarter’s total and you may get a number. Ask again and...

Using ADL and Agents to Describe and Govern Your Data

Fri, April 30

Architecture & Distributed SystemsData & Intelligent SystemsAI-Assisted Software Engineering

Many teams treat a system’s physical architecture and data architecture as separate concerns, even though a system’s data topology can directly influence its ar...

New Languages for the New World

Fri, April 30

Backend Engineering & RuntimesAI-Native Software

Every decade or so, new programming languages emerge and generate interest. The ones that gain traction often address specific problems or provide targeted solu...

Trusted Data for AI: Lineage, Freshness, and Governed Retrieval

Fri, April 30

Data & Intelligent SystemsAI-Native Software

An AI system can retrieve relevant data and still produce an indefensible answer. The source may be stale, outside the user’s entitlement, derived through an un...

The Crossroads: Augment or Replace

Fri, April 30

AI-Assisted Software Engineering

The headlines paint a picture of an inevitable future, accelerating towards humanity at a dizzying pace. It can feel as though we have been swept up in the curr...

How To Teach Your Agents About Architecture

Fri, April 30

AI-Assisted Software EngineeringArchitecture & Distributed Systems

Neal Ford

Agentic AI introduces new possibilities in software architecture, including the ability to work towards a solution when deterministic constraints exist. As deve...

Leading a Team of Developers Using AI

Fri, April 30

AI-Assisted Software Engineering

As AI becomes more prominent in software development, engineering leaders face a growing set of questions. How quickly should teams adopt it? Where can it be us...

Contain the Blast Radius of AI-Accelerated Attacks

Fri, April 30

Reliability, Observability & SecurityPlatform Engineering & DevOpsAI-Native Software

AI-assisted offensive tools can accelerate reconnaissance, vulnerability discovery, and attack chaining. At the same time, AI applications create new paths thro...

Operating Agentic Systems in Production: Evaluation, Observability, and Runtime Control

Fri, April 30

Reliability, Observability & SecurityAI-Native Software

Nicole Koenigstein

Once agentic systems reach production, the model quickly stops being the hardest engineering problem. The greater challenge is understanding why the system made...

When AI Tests, Systems, and Specifications Disagree

Fri, April 30

AI-Assisted Software EngineeringReliability, Observability & SecurityPlatform Engineering & DevOps

Peter Thomas

When the same AI influences both code and tests, a green build is no longer sufficient evidence that software is correct. The code and its tests can reinforce t...

The C4 Model, Structurizr vNext, and AI

Fri, April 30

Architecture & Distributed SystemsAI-Assisted Software Engineering

Simon Brown

The C4 model gives engineering teams a shared language for describing software architecture at different levels of abstraction, with Structurizr as its referenc...

Effect-Oriented Programming

Fri, April 30

Backend Engineering & Runtimes