Room

Agenda

Start choosing which sessions you’d like to attend at DevOps Dreamin’.

Day 1 -

Let AI carry the burden of drafting release notes

Sham Haque (CRM Tech Lead, ClearBank)

The session will take you on a journey of discovery and enlightenment, showing what’s possible with a few short prompts. We will navigate through creating an agent, discovering and connecting to reliable source data, through to refining the output to something you can share with your most critical stakeholders.

 –  AM

Breakout 2

Build Salesforce conventions that embrace change

Beech Horn (Technology Engagement Manager and Architect, Banham Patent Locks Ltd. ) and Tom Rook (Customer Product Owner, Banham Patent Locks Ltd)

Many conventions are not chosen intentionally or reinforced globally, which leaves your team (including agents) with overhead when making choices. If you have inherited an org full of conventions nobody can explain, or watched a standard erode the moment its author left, you already know the problem. Join Tom and Beech as they identify existing conventions, choose goals, weigh options, use decision frameworks, enforce standards, and ultimately allow for change. Follow our real world story from a patchwork of external contractor implementations to guidance, governance, and grief-free progress.

 –  AM

Breakout 1

Resilient Salesforce: building backup, recovery and disaster readiness into your DevOps strategy

Silvia Denaro (Senior Salesforce Manager, Hanson Wade)

WIP

 –  AM

Keynote Room

From tribal knowledge to repeatable success: writing SOPs that empower Salesforce delivery teams

Nathaniel Sombu (Salesforce Consultant, 3Strides Cloud Consulting)

Salesforce delivery teams often rely on undocumented processes and individual expertise, leading to inconsistent delivery, avoidable rework, and challenges when onboarding new team members.This session explores how to create practical, actionable Standard Operating Procedures (SOPs) that turn individual knowledge into repeatable processes. Attendees will learn how to structure effective SOPs, apply them across the Salesforce delivery lifecycle, and embed them into everyday workflows. Through real-world examples and practical techniques, participants will discover how well-designed SOPs can improve delivery quality, streamline collaboration, accelerate onboarding, and support scalable, consistent project delivery.

 –  AM

Keynote Room

Can you hear me now? Putting a voice Agent through a real release pipeline

Sajid Nadeem (Forward Deployed Engineer Director, Salesforce)

Testing a chatbot is hard. A voice agent adds latency, barge-in, accents, background noise and telephony codecs. This session lays out a layered test strategy for an Agentforce voice agent and wires it into a DevOps pipeline: Apex and Flow tests for actions, Testing Center suites via CLI/API for routing and outcomes, simulated spoken callers with Okareo for audio realism, and where end-to-end robotic tests fit. See how each layer gates promotion, how to set pass thresholds for non-deterministic results, and how production failures become tomorrow’s regression tests.

 – 

Breakout 1

Shipping quick, failing fast

Iz Wright (Product Manager, Gearset)

The biggest waste of product effort often isn't a sloppy feature. It's a polished, well-built one that solves a problem nobody actually had. In this talk, Iz Wright, Product Manager at Gearset, walks through the process behind shipping a customer-validated feature. From uncovering the "real" problem with customers by applying the lessons of Rob Fitzpatrick's The Mom Test to ask better questions, to building a rapid, shareable prototype that lets users react and interact rather than imagine. Iz shows how shipping the smallest possible slice creates space to gather real usage data, iterate fast, and even spot Go-To-Market opportunities hiding in the data.

 – 

Breakout 2

Case Studies: successful implementations of DevOps processes to reduce deployment risk

Christopher Ramm (DCX CTO Germany and Salesforce CTA, Capgemini) and Sebastiano Schwarz (Salesforce CTO Germany, Capgemini)

UI tests are the stage most Salesforce teams skip in their pipeline: too slow to write, too brittle to survive three releases a year. In this session, we share how we bring AI-generated E2E tests into the DevOps lifecycle of a seven-year-old enterprise org. Using UTAM, Salesforce’s open-source page object model, an AI agent turns user stories into executable tests that run headless as a quality gate in CI. We cover where UI tests belong in the pipeline, handling post-release failures, the guardrails that keep humans in control, and an honest look at what worked and what didn’t.

 – 

Keynote Room

Onboard your virtual teammate. Assign it half your tickets.

Laurence Boyce (Lead Sales Engineer, Gearset)

Building and maintaining custom skills for Salesforce development only gets you so far. For real AI productivity, teams need an agent they trust to pick up tickets and deliver the right changes.

 –  PM

Breakout 1

The hard part was never the tooling: bringing a Salesforce team to AI without lowering the quality bar

Patrick Mueller (Senior Salesforce Consultant, DIGITALL)

Our team went from "AI isn't touching my org" to using it daily, and the hard part was never the tooling. This is an honest lookback at AI adoption across a Salesforce consultancy: how we onboarded sceptics without mandating anything, and why the right tooling mattered mostly because it made people relaxed enough to try. I'll cover the guardrails we wrote, where AI-generated work actually fails in Salesforce metadata, and the review changes that caught it. You'll leave with a guideline template, an onboarding path, and the failures we'd skip next time.

 –  PM

Breakout 2

While with Agentic engineering you need even less DevOps Vendor tools than in 2025?

Nicolas Vuillamy (CTO, Cloudity)

In 2025, I explained at DevOps Dreamin why you might not need DevOps vendor tools, because open-source alternatives exists , even some with nice UIs ! In 2026, I'll explain why with the rise of Agentic Engineering, DevOps vendor tools are still not mandatory, but... open-source DevOps UIs might become useless, because we can delegate everything to coding agents! This shift is real, it's already here, and it's not going away

 –  PM

Breakout 2

Why we adopted source-driven deployments and lifecycle center (Next Gen DevOps Center)

Jose Carlos De Oliveira (Platform Manager and Independent Consultant, Survival International)

From Change Sets to Source Control, this session explores how we moved to a healthier Application Lifecycle Management (ALM) process by implementing source control and a staged, quality-gated pipeline using Salesforce’s native Lifecycle Center (previously known as DevOps Center). We start by understanding the foundations and fundamental purpose of sandboxes and isolated environments, why they are essential, and how source control helps address overlapping of work, dependencies and environment synchronisation. We then introduce Git and the foundations of version control, before demonstrating how DevOps Center manages changes through a staged pipeline. We’ll explore its underlying logic, work promotion strategies and the practical considerations of adopting a source-driven approach. Finally, we take a deep dive into the first edition of DevOps Center as a managed package, using practical examples to demonstrate how it works and exploring the differences introduced by the Next Generation, now known as Lifecycle Center, as a native Salesforce application. Throughout the session, we’ll share lessons learned, limitations and recommendations helping to improve collaboration, deployment processes and release management.

 –  PM

Keynote Room

Beyond static analysis: building context-aware AI Agent for Salesforce code review and tech-debt remediation

Nadeem Mohammed (Salesforce DevOps Lead, ALDAR)

Traditional static code analysis is good at identifying code-quality violations, but it often lacks the context to understand how a change impacts the wider Salesforce org. And identifying technical debt is only half the battle — actually fixing it is where things get interesting. In this session, I'll explore how to take Salesforce code reviews beyond static analysis by combining Salesforce Code Analyzer (SFDX Code Analyzer v5), AI agents, and org-level dependency context to build an intelligent, context-aware code review workflow. We'll look at how metadata dependencies can help AI agents understand relationships between Salesforce components, interpret code analysis findings in context, assess potential impact, and generate actionable remediation recommendations. More importantly, we'll explore the human-in-the-loop approach, where AI proposes and validates fixes while developers retain control over approving changes. Through a practical walkthrough, you'll see how we can move from simply detecting code-quality issues to understanding, prioritizing, and actively addressing technical debt. Walk away with a practical perspective on how AgentOps can transform Salesforce metadata quality, code quality, developer productivity, and engineering governance.

 –  PM

Breakout 1

Deploy today, release tomorrow: feature flags in Salesforce

Ellie Matthewman (Lead Technical Developer, Collinson Group)

Deploy today, release tomorrow. Feature flags separate deploying code from releasing features, letting Salesforce teams merge small changes often, ship unfinished work safely and switch off problems without a rollback. Despite claims that it “can’t be done” in Salesforce, you can build flags natively using Custom Metadata, Custom Settings and Custom Permissions. This session shows how, then covers the pipeline realities: keeping flag values right across environments, testing both paths and retiring flags before they become tech debt.

 –  PM

Breakout 2

Optimize or burn out: DevOps lessons from a solo system Admin

Justyna Krajewska (Platform Owner, Stibo Systems)

When you are the sole System Administrator, every release, request, and issue competes for your time. In this session, I'll share practical DevOps lessons, common pitfalls and real-world outcomes from building a more efficient way of managing a platform as a team of one. You'll leave with actionable ideas for using DevOps practices to scale your impact, improve consistency, and spend less time on repetitive work.

 –  PM

Keynote Room

The DORA Mirage: Are there more important metrics in the AI era?

Sam Crossland (DevOps Architect, Gearset)

"AI code generation has broken traditional DORA speed metrics. When code is written instantly, ""Lead Time"" and ""Deployment Frequency"" look elite on paper, yet actual releases still stall. The bottleneck has shifted from writing code to verifying and managing it. You'll learn how to get a more holistic view of the quality of your pipeline above the headline DORA figures, taking inspiration from Value Stream Mapping. We'll focus on areas such as PR Cycle Time and SCA violation trends, along with others, giving you better insights into how your Salesforce workflow is really performing."

 –  PM

The Library

Don’t go it alone: building backup with champions

Analissa Moreno (Salesforce Release Manager)

I’ve been Superwoman, firefighter, and team janitor on release day. It doesn’t scale. When one person holds it all together, you become the single point of failure, and you burn out fast. One brain isn't meant to rule them all. What does work? Champions. They’re not a new idea, but most of us find the best ones by accident. Finding them on purpose is a real skill: spotting the strong people on your team or the savvy cross-functional stakeholders others already go to, and building real relationships so you can lean on each other when needed. That’s how change actually sticks in Salesforce DevOps, whether it’s a new release pipeline or AI entering your workflow. In this session, I’ll share how I identify champions, what they need from you, and how to make them part of how you deliver. I’ll use real examples from leading release operations for a 1,500+ user org, messy parts included. Even if you’re a solo admin, you don’t have to go it alone, and you’ll leave with something to try the next week.

 –  PM

Breakout 1

Process Automation: beyond the pipeline

Jeremy Foster (Manager of Enterprise Applications - Salesforce, The Pilot Company)

See new ways to automate your teams standards and processes. Learn tools like Gearset Pipeline Status Updates, GitHub Actions, Branch Protections, Rulesets, and Jira Automation.

 –  PM

The Library

Increasing Developers productivity by viewing code repositories KPIs via Open Source Tools

Syed Usman Ahmad (Developer Advocate, Grafana Labs)

Whether you are just starting your journey as a DevOps Engineer or already an expert SRE, you need to use Git (e.g. GitHub, GtlLab, etc.) to maintain your code and track issues, PRs, commits, etc. You start with a few repositories but then it becomes difficult to get visibility into what's going on in a complex app when it's scattered across 20 different GitHub repos.In this talk, we will demonstrate an example of how to monitor your GitHub repo using the GitHub native API as a plugin that allows you to represent data e.g. open/close issues, Pull n Merge requests, and other various statuses visually in Grafana dashboards and gives a dynamic and interactive view to monitoring for a more adaptable and user-centric dashboard experience either for your personal projects or for the entire team. Later, we see more advanced features to get better observability. It will be an introduction to the Grafana Dashboards, and also an excellent opportunity to learn more about utilizing advanced visualization options to effectively represent complex data. Join us to learn more about Grafana dashboards, community contributions and share your feedback and suggestions!

 –  PM

Breakout 2

Unlock your team’s inner animals

Tracy Bingham (Founder, Lotus Experts Ltd ) and Danait Haile (Salesforce Marketing Cloud Specialist & Village Hotels)

Ever feel like your Salesforce DevOps team is speaking different languages even though you're all on the same org? That's because you are — just with very different natural instincts. In this fun fast paced session, we’ll look at how your natural communication style may or may not get what you need from other members of your team or your stakeholders. You’ll discover why some members charge at a sandbox refresh while others will first consider all the impacts and dependencies. Learn practical ways to flex your style for maximum impact adapting your communication to inspire each personality type, resulting in fewer conflicts, smoother releases, and a whole lot more “roar”.

 –  PM

Keynote Room

Design a DevOps Strategy for Multi-Org Implementations

Sebastiano Schwarz (Salesforce CTO Germany, Capgemini)

Salesforce DevOps falls apart in multi-org scenarios. Most projects therefore avoid multi-org strategies, even when valid reasons exist - yet most DevOps guidance assumes a single org. Teams adopting multi-org architectures quickly hit hard questions: template vs. local customisation, package-based vs. source-driven deployments, how to organise teams across orgs, and how to stop a central deployment from overwriting local metadata. This session addresses: A lightweight decision framework for when multi-org is justified, then the DevOps patterns that make it work: repository strategies, branching models, package boundaries, and governance guardrails. Built for architects and technical leads making these calls now.

 –  PM

Breakout 1

7,000+ deployments with Gearset: from release risk to release confidence

Sikandar Pathan (Senior Salesforce Developer, EVERSANA)

190 production releases, 7,000+ deployments through the pipeline, 15 sandboxes, and 700+ back-sync deployments to lower environments. These numbers are not just metrics for us, they represent how quickly our teams have been able to deliver. Working in the healthcare industry means that the changes we deliver can directly impact patient-facing programs and business operations. We needed an enterprise-level DevOps solution that could handle complex deployments efficiently while keeping multiple environments aligned. Before Gearset, we used to spend a significant part of December refreshing sandboxes to bring environments back in sync. Each healthcare program had its own custom changes, making environment management and back-sync a constant challenge. Gearset changed that approach by making back-sync an integral part of our release process. We could move changes back to lower environments without depending on manual refresh cycles. In this session, I'll share our real-world journey, the roadblocks we faced, how we addressed deployment risks, managed complex metadata and merge conflicts, and introduced continuous testing and release governance. Most importantly, I'll share what worked, what didn't, and how we built a faster, more predictable, and reliable Salesforce release process.

 –  PM

Breakout 2

Automate Salesforce DevOps: click, deploy, done

Tom Bassett (Senior Solutions Architect, Vera Solutions)

Still deploying with change sets? This session is a practical guide for admin-led teams moving from reactive releases to a trusted, automated pipeline. We cover the core toolkit (source control, work management, deployment tools, backups and test automation), how to design a pipeline from feature sandbox through integration, UAT and staging to production, and a four-stage maturity model: Reactive, Controlled, Trusted and Automated. You'll leave knowing where your team sits today and the next small step to reduce deployment risk and build trust in every release.

 –  PM

Keynote Room

A product journey with DevOps

John Belo (Senior Director, Product Management, Salesforce)

Learn how the agentic era is reshaping Salesforce's DevOps tooling and continuing a journey that started 20+ years ago with the ability for customers to customize the platform.

 –  PM

Breakout 1

Scaling a team faster than its habits: DevOps lessons from growing an engineering organisation in the age of AI

Rimon Hanna (CEO, Dutopilot)

AI agents let small teams ship like teams ten times their size, but review, testing and planning habits still grow at human speed. We run agents that write most of our code, open pull requests, test their own work, deploy to a Salesforce org, and validate the results. This talk covers what broke and what we changed: stricter tests, edge-case checks, rules in the repo that people and agents both follow, independent AI reviews, and CI costs kept in check.

 –  PM

Breakout 1

From clicks to commits: a declarative Developer’s guide to AI without losing track

Scott Reynolds (Founder, SaaS Success)

AI tools make it easy for declarative developers to build metadata, Apex and Lightning Web Components. It's just as easy to lose track: work scattered across chat threads, local files out of step with the org, and two people overwriting each other's changes. I'm a declarative developer who moved to Claude Code, SFDX projects and GitHub. I'll share what went wrong and the habits that fixed it: one source of truth, reading the org before writing, AI in sandboxes only, testing as a non-admin, and holding fixes for human review. You'll leave with a starter workflow to adopt straight away.

 –  PM

Keynote Room

Waterfall is good, actually (when AI makes it fast)

Aidan Harding (Principal Architect, Aquiva Labs)

Agile made “waterfall” a dirty word: if you discover you built the wrong thing after months of work, fixing it is expensive. Yet planning and designing before building has always been appealing. With AI, you can speed-run waterfall and get feedback in days. The structure makes important decisions intentional rather than whatever the AI happened to choose. I’ll describe how we used a spec-driven development lifecycle to deliver client projects with AI, keeping the engineering rigorous and the team aligned.

 –  PM

Breakout 2