The Real Impact of the AI-DLC
David Anderson | Author: The Value Flywheel Effect
The way we build software is changing faster than most organisations can adapt. What Dave Anderson calls the AI Development Lifecycle (AIDLC) is at the centre of that change. This isn't about bolting new tools onto existing workflows. It's a fundamental shift in how teams think about modernisation, developer experience, and delivery at scale.
Dave Anderson (author of The Value Flywheel Effect) gets specific about what that shift actually means in practice. What does a healthy engineering culture look like when AI is embedded across the stack?
How do AWS services fit into a coherent AI strategy? How do you bring your team through change without losing what makes them effective? And how do you make sure you're growing with the change, not being left behind by it?
You'll leave with a clear mental model of where AI genuinely adds value, where the hype outpaces the reality, and concrete steps you can take back to your organisation on Monday morning.
Level L100
David Anderson
10x or 0x: Engineering in the Age of Unpredictable Agentic AI
Angela Timofte | Strategic Technical Advisor | ex-VP Global Engineering: AI Trustpilot |
Founder ATim Advisory
A year ago you had a feel for how long building things took. You could sit in sprint planning and give an estimate with reasonable confidence. Today the uncertainty of estimation is higher than ever.
AI has broken engineering estimation in a specific, uncomfortable way. It doesn't make everything faster.
It makes some things dramatically faster and others unexpectedly slower, and you won't know which one you're getting until you're already in it. Sit down with a task and you're genuinely not sure if you'll be done in two hours or two weeks.
This session is a bottom-up, honest account of real engineering experience with these tools. We'll walk through concrete examples of where AI delivered a genuine 10x, and where it delivered a confident, expensive, time-destroying negative.
We'll get specific about the AWS production realities that demos skip over: Bedrock quota limits that ambush you at scale, ThrottlingExceptions under load, latency unpredictability that makes SLAs uncomfortable, and token costs that turn elegant architectures into difficult conversations. These are the things you discover after the demo.
You won't leave with a tidy framework. You'll leave with sharper instincts, honest language to use with your team, and a clearer picture of the new engineering territory we're all navigating whether we admit it or not.
Angela Timofte
Event-driven architecture, the hard parts
Yan Cui | Independent Consultant: theburningmonk.com
In this talk, let's explore the hard parts of building and operating event-driven architectures in practice, including:
-
End-to-end observability as events hop across services
-
Testing strategies within bounded contexts
-
Evolving the event schemas without breaking consumers
-
Catching integration problems early
-
Handling errors and ensuring Idempotency
-
Documentation and Governance
If you're working with event-driven architectures, come learn how to avoid these common (and painful) pitfalls.
Level L300
Yan Cui
Agentic Microservices: The Next Evolution of Microservices Architecture
Matheus Guimaraes | Senior Developer Advocate: AWS
Microservices architecture is evolving. AI agents are reshaping how backend systems are built and consumed. Discover Agentic Microservices, a new evolution of microservices architecture, along with emerging patterns such as Microservices as Tools (MAT) and Agentic Monoliths. We also explore why serverless is a natural fit for agentic architectures and watch an agentic microservice in action.
Level L300
Matheus Guimaraes
Paving the Way: Following Golden Paths on our journey from Monolith to Modern Architecture
Andrew Worden-Fitzpatrick | Staff Software Engineer: Perk
James Butherway | Staff DevOps Engineer: Perk
At Perk, we move fast. We are a Cloud and AI-native company with 380+ engineers spread across 6 global hubs and 6 time zones, autonomy is in our DNA. But with total squad autonomy comes a hidden tax: architectural fragmentation. How do you maintain "contextual consistency" in your software when dozens of independent teams are all building at once?
Our journey didn't start with a "grand rewrite." We began with a robust Django monolith that powered our early success. In this talk, we share our pragmatic approach to evolution: we aren't killing the monolith for the sake of it - we are "strangling" the parts that slow us down while keeping the parts that still provide value.
To bridge the gap between autonomy and consistency, we developed our Golden Paths. We’ll deep-dive into the theory of paved roads and demonstrate tk-create - our internal engine designed to serve production-ready AWS patterns. We’ll show how we use these paths to bake in security, observability, and AWS best practices by default, allowing a squad in Birmingham to build with the same "Perk DNA" as a squad in Barcelona, Berlin or Boston.
Key Takeaways:
-
The Autonomy Paradox: Balancing developer freedom with the need for global architectural consistency.
-
Pragmatic Migration: An example of services being extracted from the monolith and what we aim to leave behind.
-
Standardizing the "New World": How we define Golden Paths to provision DynamoDB, S3, SNS, SQS, APIGateway etc for our Lambda and ECS services.
-
Dealing with exceptions: How we quickly adopt AWS services to stop our patterns becoming a “Golden Cage”
-
tk-create & Tooling: Scaling infrastructure patterns across 380+ engineers without becoming a bottleneck and, looking to future, how AI can help us evolve our practises even further.
Level L300
Andrew Worden-Fitzpatrick
James Butherway
AWS Bedrock for serverless Intelligent Document Processing at DVSA
Shaun Hare | Principal Developer: DVSA
This talk will give a practical demonstration of the work we are doing build an intelligent document processing pipeline, using AWS textract, AWS Bedrock Data Automation and AWS Bedrock for LLM Model invocation. It will go through the why we want to do this, the architecture we developed and the challenges and results.
Level L200
Shaun Hare
Building a real-time voice agent that feels conversational, not transactional
Matthew Wilson | Distinguished Software Engineer: Instil
Niall Keys | Software Engineer at Instil | AI
In 2025 we were tasked with building a real-time voice assistant for on-the-ground sales reps working in global Pharma. They were struggling to keep accurate notes after meetings, capture outcomes from conversations and update their CRM often doing this admin work in their own personal time.
In this talk we share the story of building a production-grade voice agent on AWS using Amazon Nova Sonic 2.
This application had a measurable real-world impact on work–life balance of the sales reps. However building something that feels conversational is very different from stitching together AI services.
We’ll cover the hard lessons learnt from building real-time voice applications:
- Keeping an agent on task over multi-turn conversations.
- Managing context windows.
- Handling turn detection in noisy, real-world environments.
- Architecting for conversations longer than Lambda’s 15 minute timeout.
We’ll also explore why evals and observability with CloudWatch become mission-critical when prompt tweaks and model swaps can subtly degrade behaviour in ways traditional monitoring won’t catch.
This session is for engineers building AI systems that need to feel conversational, not transactional and who are discovering that “serverless” sometimes means knowing when not to use Lambda and look at ECS Fargate instead.
Level L300
Matthew Wilson
Niall Keys
Context Over Code: Why AI-DLC Is an Organisational Transformation, Not a Productivity Hack
Matt Houghton | AI, Data and Analytics Architect at CDL Software
AI coding assistants are everywhere, but faster code generation isn't the problem most teams need to solve. At CDL, we rolled out Amazon Q Developer to 150+ engineers and saw real results — but we quickly realised the bigger opportunity wasn't in the tools, it was in rethinking how we work.
This talk covers our journey from AI code companion to AI-Driven Development Lifecycle (AI-DLC): how we trialled and scaled AI tooling, why we built a Context Store to give AI access to decades of institutional knowledge, and what it actually takes to make AI-native development work in practice. We'll cover steering, semantic retrieval with Bedrock Knowledge Bases and MCP, verification debt, and why the developers who thrive aren't the fastest coders — they're the ones who curate context and own quality.
Level L200
Matt Houghton
From pilot to production: architecture patterns for governed GenAI on AWS
Satyen Fakey | Co-Founder & CTO of Unloq®
Mid-market and Enterprise companies aren’t short on GenAI demos. They’re short on trusted systems that turn insight into execution, stay within policy, and prove ROI. In this session, I’ll share how we’re building Strategic Intelligence on AWS: a governed decision control plane that converts KPIs, objectives, internal docs, and signals into approved actions, then records a measurable “impact receipt” each time.
We’ll cover a practical blueprint that balances tech and commercial outcomes:
• Grounded context with provenance, so outputs can be trusted
• Decision workflows using FastAPI, with AWS Lambda for event-driven steps and automation
• RDS Postgres + pgvector retrieval for traceability
• Policy-safe model and tool routing, including Bedrock where it fits
• Human-in-command approvals that create an audit trail
• An Impact Ledger, tracking expected vs realised outcomes over time
Live demo
A short end-to-end demo showing one workflow in action: decision, approval, and an Impact Ledger entry, plus what you measure after deployment.
Audience takeaway
A reusable approach for shipping strategy-grade AI on AWS that leadership can trust: faster decisions, safer execution, and measurable impact.
Level L200
Satyen Fakey
Next-Generation Resilience Testing and Disaster Recovery with AI
Sudha Arumugam | Solutions Architect, AWS
Traditional disaster recovery strategies often fall short when addressing the intricate, evolving characteristics of contemporary cloud infrastructures, creating vulnerabilities in system resilience and regulatory compliance. Explore how AI-driven capabilities can strengthen resilience and disaster recovery on AWS. This methodology connects infrastructure intelligence with application-level validation, facilitating more robust disaster recovery readiness. You'll discover how to harness Large Language Models (LLMs) alongside AWS Resilience Hub and AWS Systems Manager to transform testing approaches, evaluate infrastructure configurations, and produce customized AWS Fault Injection Service experiments and recovery procedures. Gain hands-on insights into automated test creation using templates and master the art of prompt engineering.
Level L300
Sudha Arumugam
Building Secure and Efficient SaaS Platforms on AWS Serverless
Guilherme Dalla Rosa | CTO at MerCloud
Let's go on a journey through the world of multi-tenant architectures on AWS using serverless technologies. In this talk, we will uncover the key aspects of multi-tenancy, including security, tenant isolation, and performance. We will learn how to utilise Cognito for authentication, DynamoDB to store millions of tenant-partitioned records and lambda for compute. We will also explore different deployment models and their tradeoffs, and, finally, we will learn how to implement policy-based isolation with IAM to keep our execution context tied to one specific tenant and avoid data leakage. By the end of this talk, you will feel more confident building SaaS applications on AWS with serverless technologies and you will have learned some of the many insights that come from the AWS Well-Architected SaaS Lens.
Level L300
Guilherme Dalla Rosa
Rethinking Microservices: Lessons from Refactoring to AWS Lambda Durable Functions
Matheus das Mercês | Senior Cloud Engineer at PostNL
For more than a decade, software teams have been balancing the trade-offs between monoliths and microservices on AWS. From simple Lambdaliths to complex design patterns to achieve microservices benefits, this debate has been anything but settled.
At AWS re:Invent, AWS introduced Lambda Durable Functions with an interesting premise: build like a monolith, deploy as microservices. By bringing durable execution to Lambda Functions, AWS is challenging assumptions about how serverless applications should be designed.
In this talk, I will walk through a real-world refactor of a multi-step data pipeline at PostNL, migrating from AWS Step Functions to AWS Lambda Durable Functions and still respecting the application's design pattern. We’ll explore what worked, what didn’t, and the trade-offs we encountered along the way.
When you walk out of this session, you will understand how Lambda Durable Functions work under the hood, when they are a better fit than state machines, the trade-offs they introduce, and how they influence microservice design decisions in practice, or whether this marks the end of the monolith vs. microservices debate as we know it.
Level L300
Matheus das Mercês
Building an LLM Inference stack on AWS
Paco Gonzalez | Snr Specialist Solutions Architect, Accelerated Computing at AWS
Ben McConnell | DevOps Manager at Luminance
AWS provides a multi-layer LLM inference stack to address customers' diverse inference needs. This session focuses on building an inference stack on Amazon EKS. Learn a systematic approach for sizing GPU infrastructure by translating user-centric metrics into system requirements, accounting for workload patterns and traffic bursts. We dive deep into the core building blocks of the stack, including Karpenter for implementing effective capacity management, and explore various infrastructure and inference optimisations. This session features Luminance, a leader in Enterprise Legal AI, who will share their experience evaluating multiple approaches on AWS to scale their LLM capabilities for advanced document analysis.
Orchestrating a workflow in Lambda
Chris Dobson | Senior engineer @ Trustpilot & AWS Community Builder
There’s a new kid on the block in the world of workflow orchestration, the Lambda Durable Function, which allows an orchestration to be created as a single Lambda Function which can run for up to a year.
-
How do you build an orchestration?
-
How much will it cost?
-
What pitfalls should you look out for?
-
Can it do everything you need?
-
How do you test or debug it?
Are just some of the questions I had about this new extensions to Lambda.
This talk will take an existing production workflow and look at how it could be built as a Lambda Durable Function from start to finish. I’ll implement the complete workflow, look at how the checkpoint/replay execution model affects this implementation, test and debug the workflow, and look at versioning all through a number of demos which have helped me answer those questions.
Level L200
