A welcome to LeadDev Berlin 2024 from the host, Pat Kua.

What does “innovation” really mean, and what are the conditions that enable it? The book “Collective Genius” highlights that Leaders of innovation need to make sure they collaborate, engage in discovery-driven learning, and make integrative decisions. Additionally, embracing “paradoxes” in every situation will help you approach problems with a “both/and” mentality. Learn from our struggle in Games Engineering to create this environment and how we embraced these tensions to unlock innovation for Netflix Games.

In the GenAI era, we’re outsourcing more than tasks – we’re outsourcing trust. This talk explores how to lead creatively and confidently when everything starts with a prompt.

This abstract has been created without relying on any GenAI tool. This was hard for me, I’m not a story teller, not a writer, I barely had conviction in what I had to say before GenAI came along. It’s been an incredible tool for me and countless others, It feels like my productivity shot through the roof, I had access to this wealth of knowledge, I felt like I had a voice – a seat on the table.

Over the last year, I’ve felt this voice get quieter.

In chasing speed, polish, accuracy, I’ve started to default to the tool – sometimes even before I’ve formed my own thought. “Hold on, let me prompt that”. For many of us, it’s become a way to delay decision-making, outsource confidence, and avoid the discomfort of trusting our own thinking. This talk is about exploring that tension: between the incredible power GenAI gives us, and the quiet erosion of the narrative conviction it can cause. For Engineering Leaders, this isn’t theoretical. We lead people. We make bets. We communicate vision. And when our sense of authority is blurred—by too many prompts, too much polish, or too much second-guessing – we risk leading from pre-defined templates instead of truth.

This talk invites us to take a pause, to sit with that moment where the answers aren’t clear yet, that thought isn’t fully refined yet. In a world where everyone has this tool, our advantage as engineering leaders is still the experience and perspectives we bring as individuals.

Key takeaways

• Recognize early signs of Gen-AI induced confidence erosion
• Rebuild your narrative and inner-clarity to enable high-stakes decisions as an Engineering Leader
• Discern when to prompt, and when to lead from conviction
• Apply simple techniques to model original thought and independent judgment

As leaders we have been tasked to go all-out with AI: both in terms of our products and also our productivity. This talk outlines what you should be thinking of doing for your team, your product, and, importantly, yourself with AI.

The hype surrounding Large Language Models (LLMs) has reached a fever pitch, making it challenging for technology leaders to discern actionable strategies from speculative promises. This talk cuts through the noise, offering a pragmatic operator’s perspective on how to genuinely leverage LLMs within your organisation.

We’ll cover specific, real-world applications for boosting your engineering team’s output, from automating mundane tasks to enhancing code quality and fostering rapid prototyping. You’ll learn how to identify high-impact product features powered by LLMs that deliver quick wins, along with practical considerations for data privacy, model selection, and deployment.

Crucially, we’ll discuss how you, as a leader, can personally leverage AI for better decision-making, thinking and strategic planning, ensuring you leave with concrete steps to implement and see immediate value from AI across your team, product, and leadership approach.

Key takeaways

• Develop a clear, actionable AI strategy for your team: Learn how to assess the readiness of your engineering function for LLM adoption and identify practical tools and workflows to boost productivity and innovation without overwhelming your team.
• Identify impactful LLM applications for your product: Understand methodologies for pinpointing genuine problems LLMs can solve within your product roadmap, moving beyond surface-level integrations to create meaningful user value.
• Navigate the personal leadership shift with AI: Gain insights into how your role as a technology leader evolves with the rise of AI, including strategies for continuous learning, ethical considerations, and fostering a culture of responsible AI experimentation.

What if audits weren’t painful, but automatic? Discover how to turn compliance into code, eliminate fire drills, and run a thousand audits a day, without breaking a sweat.

We deploy code a thousand times a day, yet when it comes to compliance we still use spreadsheets like it is 1985. Endless paperwork. Manual reviews. Last-minute fire drills. We’ve all been there.

It does not have to be this way.

This talk challenges that status quo. It introduces a practical, forward-thinking approach where compliance is automated, seamlessly integrated into CI/CD pipelines, and capable of detecting risks in real time. Where engineers get instant feedback without switching context. Where compliance is just part of the dev flow, not a blocker at the end of it.

So, what does painless compliance actually look like?

First, we need to treat compliance the same way we treat customer requirements: as features. Not as 400-page PDFs filled with legalese, but as clear, structured requirements defined in our own language, tailored to our systems. When we do this, we unlock real conversations, reduce ambiguity, and make compliance understandable and actionable.

Next comes the magic: automated validation. It’s not enough to say we’ve implemented a control, we need to prove it, continuously. If storage encryption is required, we don’t just enable it. We generate dashboards that prove it’s working. We build reports that our compliance teams can take straight into the audit, with no disruption to engineering.

And we monitor it. If something goes out of compliance, be it unencrypted storage or misconfigured access, we get alerted immediately. No surprises, no scrambling before an audit.

And here’s the best part: if you’re in the cloud, you already have everything you need to do this today.

A thousand audits a day. One click. Zero drama. Not only is it possible, it’s the future. Let’s build it.

Key takeaways

• Compliance can be automated
• Treat compliance as a product requirement
• Prove, don’t just implement
• Empower engineers with instant feedback
• You already have the tools”

Growth isn’t always vertical, and forcing people up the ladder often pushes them out. This talk challenges the frameworks we use to define success and shows how to build teams that grow without leaving.

We’ve convinced ourselves that growth = promotion. That career success means moving “up.” That to retain people, we must level them – whether they want to or not.

But what if someone doesn’t want to climb? What if their best work, their most fulfilled self, exists in a role without a title change?
In this lightning talk, I’ll expose the growth frameworks we use (and often weaponize):

• The Career Ladder = deeply hierarchical → Promotions-as-growth, limited mobility → can’t be redeemed Burn it
• The Dual Track Framework = rigid → Hard to switch, unequal prestige → good if we Redesign it
• Competency Matrix = oversimplifies → Used as gatekeeping, checklisted → Reclaim it with care

These systems were built for order, not for humans. And they often punish the curious, the multi-passionate, and the quiet high performers.
I’ll share real stories of team members who thrived only after we ditched the “climb” mindset, and started treating careers like something to explore, not just escalate.

You’ll walk away with permission to rethink what growth really looks like, and how to build systems that support it without forcing everyone up.

Key takeaways

• Not all growth is vertical, and that’s okay.
• Promotion isn’t always retention.
• Competency frameworks often punish real potential.
• You can support growth without offering a new title.
• Great teams make space for exploration, not just elevation.

Sponsored by

How to use both evals and experiments in your AI development lifecycle to avoid costly mistakes and ship more successful projects.

AI initiatives are anything but a slam dunk: projects can be expensive, hard to measure, and face many failures along the way. The key to success lies in building the capability to learn quickly and fail cheaply. By going beyond upstream, qualitative evals and incorporating downstream, quantitative experiments, teams can shorten feedback loops and allow for rapid course correction. In this talk, Datadog Senior Technical Advocate Ryan Lucht will show teams how to leverage both approaches and join model performance with business metrics to ship more successful AI initiatives.

Key takeaways

• How to handle the low success rate of AI initiatives by failing cheaply
• The different purposes of evals and experiments – and why you need both
• Using error analysis to define and measure evaluators
• Running experiments at scale with basic and advanced A/B testing approaches for AI development

Developing a clear target architecture will take your technical leadership to the next level by improving collaboration, increasing efficiency, speeding up execution and allowing you to make data-driven decisions while building to scale.

Can you imagine if construction workers built a skyscraper using notes scribbled on random bits of paper? Imagine how they would coordinate the plumbing, electricity, floor plans…Yikes! Yet, this is sadly what most tech systems look like – patched up to “make things work”, always fixing “the previous person’s code” and constantly bumping into chunks of code which make absolutely no sense but you’re too scared to remove because you’re pretty sure it does something important somehow.

As the founding engineer of the Cash Flow domain in Pleo, I was privileged enough to play a key role in architecting a whole subsection of the company’s backend ecosystem while building up a team that would take it to new heights. In this talk, we’ll walk through this case study which covers the wins and fails of our journey and ultimately highlights the importance of having a target architecture and how it has helped inform decision-making (technical and non-technical), increase efficiency, improve collaboration and speed up execution while building to scale.

Key takeaways

• What is a target architecture diagram and how to create it for different stakeholders
• How target architecture informs decision-making (technical and non-technical), increases efficiency and speeds up execution
• How to grow domains and teams based on target architecture

Tech NGO partner

CodeYourFuture helps people change their lives by training exceptional early-career engineers without prior tech experience straight into advanced roles such as Platform SRE at Slack. Come learn how, be wowed by our graduates’ achievements, and talk about how you can hire amazing top talent more sustainably.

CodeYourFuture trains exceptional talent for advanced roles from under-served communities. We help high-potential people in need to change their lives by getting good jobs in tech.

We’ve trained people with no software background straight into high-achieving roles such as platform SREs at Slack, typically positions reserved for high-tenure, highly competitive, highly expensive candidates. We’ll share how we achieve this with world-class high-touch training, and how we can help you hire excellent talent more sustainably.

We’ll cover:

• The kinds of communities we work with who you may not currently see represented in your staff.
• How our course prepares people to join your teams better than bootcamps or universities would.
• How we can help build sustainable and predictable hiring pipelines.
• How we can help your teams up-skill their mentoring

But most of all, we’ll brag about some of the truly impressive things our graduates have done.

If you think managing managers is just a bigger version of managing individual contributors, think again.

The strategies that launched your engineering career and got you into leadership won’t carry you to the next level. This talk challenges your assumptions and reveals what it truly takes to shift from doing it yourself to enabling others to do it better.

Sponsored by

A deep dive into the high-performance, distributed architecture and machine learning optimizations Cloudflare uses to detect malicious attacks at a global scale with sub-millisecond latencies.

In a world where the scale of internet traffic is constantly growing, how do you protect against malicious attacks when you’re handling over 84 million requests per second? This talk will pull back the curtain on Cloudflare’s approach to high-throughput, low-latency threat detection. We’ll explore our distributed architecture that spans data centers worldwide, and how we leverage Rust for extreme efficiency and resource optimization, where every CPU cycle and byte of memory counts. We will delve into the smart processing techniques that are critical to our success, including advanced caching strategies and hardware tuning. A significant portion of the talk will focus on the machine learning models at the core of our Web Application Firewall (WAF), and the extensive optimizations we’ve implemented, from SIMD and TensorFlow Lite upgrades to LRU caching, to make them incredibly fast. We will also touch on our innovative data generation and sampling strategies that are key to training accurate and resilient AI models.

Key takeaways

• Architecting for Scale: Learn how to design and build distributed systems capable of handling massive volumes of traffic with low latency.
• Performance Optimization: Discover practical techniques for optimizing software and hardware, including the benefits of using Rust and advanced caching.
• ML in Production: Understand the challenges and solutions for deploying and optimizing machine learning models in a high-stakes, real-time environment.
• Actionable Scaling Strategies: Gain insights into how to leverage serverless and distributed architectures to scale your own applications and ML algorithms.”

How I used AI tools to accelerate onboarding, understand team dynamics faster, and show up more confidently as a new engineering manager without drowning in docs or meetings.

Starting a new role as an engineering manager can be overwhelming. You’re expected to absorb technical context, understand team dynamics, and start delivering impact all within the first few months. In my most recent onboarding experience, I decided to do something different: I let AI help.

In this talk, I’ll share how I used AI tools to streamline my onboarding as a new manager. These tools didn’t just save time they helped me surface insights I might have missed, get clarity faster, and reduce the pressure of “figuring everything out” on my own.

I’ll walk through real examples of how I used AI to synthesize org charts and product documentation, explain complex systems, interpret unfamiliar code reviews and engineering standards, and improve team workflows. I’ll share how I mapped out my new engineers’ roadmap, identified gaps in our processes, and even prototyped onboarding plans all with AI’s support. I’ll also talk about what didn’t work, where human judgment was essential, and how I balanced automation with building authentic, trust-based relationships.

This isn’t about hype it’s about how AI can be a practical partner in the messy, high-context reality of engineering leadership. Whether you’re a new manager, hiring one, or just curious how AI is changing the way we work, you’ll walk away with practical ideas and honest reflections from the front lines.

Key takeaways

• How to use AI tools to accelerate understanding of org context and culture
• Pitfalls to avoid when over-relying on automation
• How AI can reduce early-stage overwhelm for new managers

A practical playbook for leading engineering teams through 2025’s budget pressures; aligning technical decisions with business goals, empowering teams, and driving impact despite constrained resources.

In 2025, engineering leaders are under pressure from all directions. Customers expect faster feature delivery and tools that deliver even greater value. Product owners and the C-suite push for innovation and speed. Meanwhile the CFO demands all this, but without increasing spend. Budgets are tighter, hiring is slower, and AI is reshaping how teams operate. And yet, expectations keep rising. This talk is a practical playbook for navigating that tension.

At StatusCake, a bootstrapped SaaS company, we’ve never had the luxury of excess. We’ve operated for over a decade with lean teams, constrained budgets, and no external funding. That’s taught us how to align teams around impact, prioritise ruthlessly, and make technical decisions that directly support the business; all without burning people out or compromising on quality.

I’m not an engineer myself, but I’ve led engineering teams for 15 years. That perspective has forced me to build trust through clarity, not code. And to translate between product, engineering, and commercial strategy with precision. In this session, I’ll share the practical lessons that emerged from leading through constant constraint, and how they apply to the challenges engineering leaders now face at scale.

This is not a just a story of surviving 2025, but it should help you create a framework for a thriving engineering team within your organisation. Helping you and your team stay focused, motivated, and strategic when resources are scarce but expectations are high.

Key takeaways

By the end of the session I hope you’ll have learned 5 things to take back to your businesses:

1. Why budget constraints are a leadership challenge, not just a financial one: Because leading through scarcity is about alignment, not spreadsheets.
2. How to empower engineers to think commercially and make smarter technical trade-offs: When engineers understand the “why,” they make better calls on the “how.”
3. Ways to align tech decisions with company-wide goals during moments of pressure: Strategy fails when tech and business drift out of sync; especially under stress.
4. How constraints can sharpen focus and increase team autonomy: Limits don’t have to reduce creativity, they can accelerate it.
5. How to reshape the role of engineering from a cost centre to a strategic driver of value: Engineering isn’t just delivery; it’s a key lever for business performance.

Think you’re stuck in your career?

You’re not.

You’re just following the wrong rules.

This talk will show you how to get noticed, grow your influence, and finally get the recognition you deserve.

Are you working hard, hitting your goals, and still not getting the recognition or growth you deserve?
You’re not alone – and you’re not broken.
You’re just playing by rules that were never designed for you.

In this talk, I’ll share the unspoken truths about career growth in tech, especially for women, introverts, and hidden leaders who feel like they’re doing everything “right” but still getting overlooked.

From my experience as a Director of Engineering and now an executive coach for women in tech, I’ll walk you through what actually moves the needle – how to build visibility, influence decisions, and shape your path without selling your soul or burning out.

You’ll get a practical and honest framework for driving your career forward on your terms – and stop waiting for permission to lead.

This isn’t another “just be confident” pep talk.
It’s a real conversation about what works, what doesn’t, and how you can start leading now – even if no one handed you the title yet.

A light, practical talk for engineering leaders on tackling self-doubt using a debugging mindset. Learn how to build real confidence with a simple framework that improves not only you, but also your team.

As engineering leaders, we’re expected to have answers, make decisions, and support our teams – even when we’re quietly questioning ourselves. Speaking up in high-stakes meetings, giving feedback, or navigating ambiguity can trigger self-doubt and quietly erode confidence over time.

In this talk, Natalia shares how imposter syndrome shows up in leadership – and how to handle it using a mindset we already know well: debugging. With humor, personal stories, and practical tools, she introduces the Confidence Debugging Framework (CDF) – a 5-step approach to identifying confidence “crashes” isolating their causes, applying targeted fixes, and building habits that stick.

You’ll learn:

• How to recognize your own confidence blockers and triggers
• Simple, proven strategies to build confidence intentionally
• Why your confidence directly impacts your team’s trust, performance, and growth

Expect a refreshing, actionable take on a topic that’s often overlooked in tech leadership – but deeply impactful. From private self-doubt to public team dynamics, this talk will leave you with tools to lead with more clarity, confidence, and self-awareness.

Because just like code, confidence isn’t perfect – but with the right mindset, you can always refactor it.

Key takeaways

• Recognize how self-doubt shows up in everyday leadership.
• Use a simple debugging mindset to approach confidence challenges.
• Discover practical habits for building confidence over time.
• Reflect on how your confidence affects your team’s trust and dynamics.

Charity’s new talk will address the incredible variance in performance among engineering teams in 2025. While the ceiling for engineering performance has increased dramatically over the past decade, only a tiny fraction of teams are actually taking advantage of this shift. Hear how your observability team can help unlock the next level in performance.

Closing thoughts from the host, Pat Kua.

If you’re inspired by a talk, want to learn more, or just want to chat about it with other folks, head to our Office Hours area where our speakers will be waiting for you.

Dedicated sessions on how to take the next step in your career for stepping into the role of a Staff Engineer.

Moderated by Phil Bennett, Experienced Engineering Leader & Former Director of Engineering at Kilo Health.

Watch or participate in free 1:1 coaching with Lena Reinhard.

If you’re inspired by a talk, want to learn more, or just want to chat about it with other folks, head to our Office Hours area, where our speakers will be waiting for you.

Dedicated sessions on how to take the next step in your career for preparing for the role of a Principal Engineer

Moderated by Phil Bennett, Experienced Engineering Leader & Former Director of Engineering at Kilo Health.

If you’re inspired by a talk, want to learn more, or just want to chat about it with other folks, head to our Office Hours area, where our speakers will be waiting for you.

Sponsored by

Whether you’re a CTO or a rising engineering leader, you’re likely under immense pressure to ship more value, faster. Yet, you know that jumping in and contributing code is unlikely to be the path to success. So, as a technical leader, how do you help your teams? They are likely dealing with increasing system complexity, siloed knowledge, and mounting technical debt. Many engineering metrics (DORA, DX Core 4, etc) can confirm there’s a problem. You should definitely talk to your people, but they have biases and will either tell you what they think you want to hear, or tell you what they have concluded is the problem. So how do you get past the gut feeling and put your skills to work to optimize your org and systems for software delivery excellence?

The answer lies in reframing the technical understanding you already have at a system level, using the strategies of systems thinking. It’s not about applying technical knowledge to solve technical problems (e.g., rewriting code), but rather paying attention to the flow of code and information to remove blockers in systems. By analyzing how code and PRs are structured, understanding how feedback flows through systems, and identifying where breakdowns occur, you can unlock the full potential of your software delivery system.
No matter how big or small your systems are, this workshop will help you develop the superpower of reading your systems’ vital signs to create environments where teams can deliver software quickly, confidently, and effectively.

What will you learn?

• How to use your technical expertise to identify hidden system issues by analyzing code patterns, PR structures, and team interactions.
• How to measure and quantify technical bottlenecks rather than relying on subjective assessments when prioritizing improvements.
• Tips for designing team structures based on evidence of what actually enables faster delivery vs. relying on intuition.

Sponsored by

Grab lunch, meet your peers and discuss core engineering challenges

Finding the right solution to complex engineering issues isn’t about isolated thinking – it’s about building your support network and leveraging the power of peer learning through diverse experiences.

Join us for a relaxed, drop-in Lunch & Learn session where women leaders in engineering will discuss key challenges, share their experiences, and learn from one another. The session will be facilitated by five experienced women engineering leaders who will guide the conversation.

Join conversations focused on

• Building strong partnerships and upward influence to drive success for you and your teams
• Preventing burnout for yourself and your teams
• Evolving your leadership approach and identifying the skills your team needs to thrive in the age of AI

Artificial Intelligence is no longer on the horizon; it’s in our IDEs, our pipelines, and our workflows. The 2025 State of AI-assisted Software Development report reveals a critical paradox: while 90% of developers now use AI and throughput is finally increasing, software delivery instability is also on the rise. How can we harness AI’s power without amplifying chaos?

The answer lies in a fundamental insight from the DORA research team: AI is an amplifier. It magnifies the strengths of high-performing organizations and the dysfunctions of struggling ones. Success with AI isn’t a tools problem—it’s a systems problem.

In this highly interactive workshop, we will move beyond the hype and dive into the data. We won’t just share the report’s findings; we will help you apply them to your own teams. We will explore and discuss some of the competencies identified in the new DORA AI Capabilities Model

You will leave this session with:

• A clear understanding of foundational capabilities that unlock AI’s true potential.
• Actionable advice to diagnose your team’s strengths and weaknesses.
• A framework to ensure your AI strategy translates into measurable improvements in performance and stability, not just localized pockets of productivity.

Sponsored by

AI has changed how code is written, but the real bottleneck is getting this code to your customers in production. Join Harness for a look at how AI is transforming the entire software delivery lifecycle, from how teams test, secure, deploy, and optimize software at scale.

We’ll unveil Harness’ latest AI innovations and show what’s now possible when intelligent workflow orchestration meets your software delivery data. Through live demos, you’ll see how Harness turns AI innovation into concrete results—accelerating delivery, improving security and reliability, and unlocking developer productivity.

Sponsored by

Explore solutions for your technical issues.

See how AI can transform incident management from chaos to clarity. In this 10-minute demo, we’ll show how incident.io’s new AI SRE autonomously diagnoses issues, assists with investigations, and empowers teams to resolve incidents faster – without replacing the human judgment that matters most.

Sponsored by

As AI-generated code accelerates development, teams face an impossible choice: ship what you haven’t fully validated, leading to buggy production and mounting tech debt, or lose all your AI productivity gains by creating validation checkpoints that feel like finding a needle in a haystack. Enter Chunk, CircleCI’s new autonomous agent (beta) that works 24/7 to fix the critical but time-consuming parts of your software delivery process that everyone knows are important but no one has time to maintain.

In this demo, we’ll focus on Chunk’s flaky test capability, demonstrating how it autonomously analyzes and fixes non-deterministic tests while learning your team’s patterns and best practices. We’ll explain why Chunk is a fundamental shift to proactive CI/CD processes that free engineers to focus on creative problem-solving and building features that differentiate your product.

Sponsored by

Explore solutions for your technical issues.

Make the most of your breaks with the demo stage sessions, where leading tech providers will showcase their innovative solutions in informative and concise presentations. Each carefully curated 10 minute demo will show the very best engineering solutions available.

Sponsored by

Prototyping is easy; production is messy. Retool shows how their customers have brought order to the chaos of app development by blending low-code, AI, and developer expertise to build robust apps that connect with real infrastructure and data.

Sponsored by

See how LaunchDarkly enables teams to deploy continuously whilst maintaining complete control over what users experience. In this demo, we’ll show how feature flags, progressive rollouts, and real-time impact measurement work together to let you move from idea to production in minutes, not weeks. Perfect for teams dealing with increased code velocity from AI tools or looking to modernise their deployment practices without adding risk.

Sponsored by

In one workflow, see how your team can write, run, and scale Playwright tests as production synthetic monitors. In 10 minutes, you’ll learn how to cut costs, boost reliability, and use AI to setup and maintain of your observability code.

Sponsored by

Instantly transforms your engineering guidelines, standards and best practices into live system-wide enforcement with Pandorian

Sponsored by

See how v0 is changing the way we prototype and build products. In this session, Leo will show how to combine v0, the Vercel AI SDK and the AI Gateway to build a fully functional, modern chatbot with agentic tool calls in just ten minutes.

Join fellow DirectorPlus leaders from underrepresented communities to connect, share experiences, and build community. This informal meetup offers space for open conversation among LGBTQIA+ and BIPOC leaders, neurodivergent and disabled professionals, women, and anyone who identifies as underrepresented at senior levels in tech.

Guiding discussion prompts will help spark meaningful exchanges about leadership, belonging, and navigating challenges as “the first” or “one of few” at your level.

If your identity or experience isn’t explicitly mentioned but you feel underrepresented, you’re warmly invited to join us.

You can also take part in the Community Representation Groups running throughout the day as part of LeadDev.

A welcome to LeadDev Berlin 2024 from the host, Pat Kua.

Hackathon host Yenny Cheung takes us behind the scenes.

Get the inside story on what unfolded at the first-ever LeadDev AI Hackathon.

You spend a lot of your workday thinking, but how much time do you spend rethinking? Failure to challenge our own assumptions can lead us to miss big opportunities. Sometimes, failure to rethink can even have deadly consequences.

Through real-world examples from NASA’s Challenger explosion to Blackberry’s failure to thrive, we’ll examine how thinking like a scientist can make us a better leader and improve team dynamic.

You’ll learn how to challenge your assumptions, encouraging you to actively seek out information that challenges your views.
You’ll learn how to inspire a learning mindset in our teams through a strong foundation of psychological safety and promoting constructive conflict.
And you’ll learn why we should replace a culture of “best practices” with a learning culture.

We’re quick to update our possessions. When our clothes are out of style, we go shopping. When our phones are old, we replace them. So why shouldn’t we update our opinions and assumptions? It’s time we stop listening to views that make us feel good and seek out views that make us think hard.

Key takeaways

• Frequently updating our own views makes us a stronger leader.
• Challenging others to open their minds and potentially change their views strengthens team bonds and allows for healthy discussions.
• Replacing “best practice” culture with a learning culture at work improves psychological safety and efficiency.

Insights from the inaugural LeadDev Engineering Performance Report, with Honeycomb. Find out how organizations measure the performance of their software and systems, the impact AI is having on that performance, and the most impactful ways to unblock performance bottlenecks.

Learn how to perform large-scale migrations and rewrites of complex systems safely. An experienced engineering leader will share his learnings from leading teams that have successfully performed multiple migrations over the past few years.

Performing large-scale migrations, rewrites, and sunsetting systems in software engineering is inevitable. They are an organic part of continuously growing, improving, and evolving the systems we work on. Teams and organizations that master these tasks successfully have a competitive advantage when it comes to adapting to changes and innovation.

This talk will share hard-earned lessons and insights on how to successfully migrate key functionalities safely and without negatively impacting users. It will first explore common reasons why teams usually need to undertake these projects. Then, it will present and assess different execution strategies, focusing in particular on different techniques that enable having incremental progress instead of a big bang switchover. Finally, it will cover how to effectively measure and communicate the progress of the migration, as well as how to build trust with stakeholders.

Key techniques discussed in this talk will include:

• Learning how to break down your problem and set up an incremental delivery plan
• Partial rollouts that provide faster feedback loops and avoid risky releases
• Dual-read approaches, on-the-fly migrations, and hard cuts
• Traffic splitting through feature flags and load balancing

The speaker will share his firsthand experience and learnings after having migrated several complex systems during the last years. Attend this session to gain valuable knowledge on ensuring your migrations are both safe and successful.

Key takeaways

• Understand common triggers for large-scale migrations and rewrites
• Learn strategies for executing migrations successfully and understand their trade-offs
• Discover techniques for planning and implementing incremental rollouts
• Learn how to build trust and effectively communicate the progress of the migration to your stakeholders
• Explore methods for traffic splitting at both application and infrastructure levels

How can senior ICs and managers strategically guide junior engineers to develop critical engineering skills amid the rise of AI coding assistants?

In a landscape where generative AI and coding assistants are increasingly prevalent, how do we ensure that junior engineers develop the critical thinking and problem-solving skills necessary to progress to senior roles?

This talk explores how senior ICs can foster a learning environment that leverages AI tools without allowing them to become crutches. We will discuss frameworks for evaluating when AI can accelerate learning versus when it might obscure skill gaps, and share case studies on effective mentorship strategies in this new paradigm.

Attendees will leave with actionable insights on balancing AI-assisted workflows with traditional engineering skill development, ensuring that emerging engineers are prepared to thrive as senior ICs in the future.

Key takeaways

• Identify key skill areas that AI tools may obscure and strategies for reinforcing them.
• Implement mentorship frameworks that balance AI use with traditional problem-solving techniques.
• Develop actionable plans to assess junior engineers’ readiness for senior roles amid AI-driven workflows.

How to harness the explosive speed of AI-driven development without losing control, compromising safety, or putting customer experience at risk.

AI is rewriting the rules of software delivery. Features that once took weeks to design and build now appear in minutes with the help of copilots. Prototypes rapidly morph into production-ready code, and the volume of changes flooding into systems feels like an unstoppable tsunami.

While this acceleration is a gift, it also introduces new risks — moving too fast without the right guardrails can compromise safety, reliability, and customer trust.

In this session, we’ll explore how leading engineering teams are evolving their delivery practices for the AI era. You’ll learn how to separate deployment from release to maintain control, and walk away with actionable practices to launch software at AI speed — safely.

Separating Deployment from Release — controlling exposure without slowing innovation.

Actionable Practices — concrete steps to launch software at AI speed, safely.

The gap between AI demo and AI product is full of hard engineering work. I’m sharing 10 years of experience involving scaling, orchestration and cost optimisation.

How much does it take to go from a shiny AI demo to a deployed application? Everyone’s excited about new AI capabilities, but few talk about the engineering work required to use ML models in production. Getting this hidden part right is what separates successful AI projects from failed ones. In my talk I’ll show what it takes to turn ML experiments into robust, scalable products.

At Chattermill we’ve spent 10 years bringing ML to production. I’ll share hard-won lessons on orchestration of a dozen self-hosted ML services. We’ve had to tame the unpredictability of LLMs at scale, bridge the velocity gap between engineering and data science, and drive down both processing time and infrastructure costs. Have you ever thought that innocent retries could cost you hundreds of dollars?

You’ll walk away with methods to iterate faster and deliver more reliably in a heterogeneous environment of data scientists and software engineers. Whether you are a tech lead, architect or manager, this talk will prepare you to ship real-world ML applications, not just experiments.

Key takeaways

• Dealing with pace differences between data science and engineering
• Methods to iterate faster with ML products
• Awareness of the hidden effort required to bring ML to production and differences between regular software projects and ML ones

No matter what your org chart says, if you’re building a lasting product, you’re in a product and tech relationship. Make it a good one.

Good means effective and efficient, but also enjoyable. That kind of partnership takes time, care, and intention.

Here are the three pillars of a strong product and tech relationship:

1. Alignment – Are you actually on the same page? Leadership styles, ways of working, and long-term goals need to match before you can make real progress.
2. Date Night – Keep the spark alive with regular check-ins, syncs, and collaborative rituals to ensure ongoing alignment — especially at the leadership level.
3. Best Friend to Moan To – No relationship is without its challenges. Have a trusted partner to vent, troubleshoot, and work through friction before it turns into conflict.

Key takeaways

• Tools and best practices for alignment and collaboration
• How to create moments of being human to develop trust and make your relationship stronger
• A laugh or two as I share stories from the messy reality at the intersection between product and tech and what we can learn from them.

What does it really mean to support people who are different and create an inclusive working environment for all neurotypes? Learn the difference between lip service and real diversity.

This talk is about the difference between paying superficial support to diversity and inclusion, versus truly understanding how different colleagues, team members, and employees have varying challenges and needs. From the small, everyday details to the bigger picture, this talk includes practical ideas and examples of how your behaviour and the support you offer can have a major impact on whether you and those around you are able to thrive.

Clare draws on years of experience supporting people with a wide range of needs as a mentor, tutor, interviewer, and line manager. She also shares her own perspective as a middle-aged autistic woman with ADHD navigating menopause. Through personal stories, she highlights the challenges she has faced and what has made a real difference.

Small actions can have a big impact.

This is a grounded and honest look at what it really means to be different, with practical advice and insights on how greater awareness can lead to meaningful change.

Key takeaways

• How to identify the needs and strengths of all your team members
• How to create a supportive working environment for people with differing needs
• What neurodivergence means
• The strengths of neurodivergent team members

A real-world story of how we integrated AI into our product development workflow, what broke, what worked, and how it’s changing our sprints, our quality, and our team habits.

How do you bring AI into your product development workflow without slowing your team down or making things worse?
At Learnn, an italian EdTech startup with over 200,000 users, we set out to do just that.

In this talk, I’ll walk you through how we introduced AI into our engineering process, not as a shiny tool, but as part of how we build, test, and deliver product. You’ll see what worked, what failed, and how AI is now helping us write code, deliver features, generate tests, improve documentation, and even speed up product discovery.

We’ll look at the impact on sprint velocity, product quality, and team dynamics with honest lessons from the trenches.

Whether you’re leading a small team or scaling a large org, this talk will give you practical ideas (and cautionary tales) on how to adopt AI in a way that’s sustainable, measurable, and team-friendly.

Key takeaways

• A practical framework to start integrating AI into your workflow.
• Key insights into what AI is good at and where it breaks.
• How AI adoption changed our sprint dynamics and team mindset.
• Strategies to introduce AI without losing control or quality.
• A real-world use case of AI’s impact on our product delivery

AI is reshaping what’s easy to solve in technical interviews, prompting us to revisit what we ask and why. This talk explores how more thoughtful, human interviews can lead to deeper signal and better hiring.AI is reshaping what’s easy to solve in technical interviews, prompting us to revisit what we ask and why. This talk explores how more thoughtful, human interviews can lead to deeper signal and better hiring.

AI is changing the landscape of technical interviews. With tools that can generate code, explanations, and solutions in seconds, many traditional formats are losing their ability to surface meaningful signal. At the same time, AI is becoming a natural part of how engineers work – so why design interviews that ignore it?

This talk explores how we can adapt interviews to this new reality. Rather than relying on questions that are easy to offload, we’ll focus on ways to uncover deeper qualities: how candidates think through ambiguity, weigh trade-offs, and communicate their decisions. It’s not about avoiding AI—it’s about designing interviews that reveal what it can’t answer for someone.

We’ll dig into practical strategies for structuring interviews around your team’s actual workflows, crafting questions that invite thoughtful discussion, and creating an environment where candidates feel comfortable showing how they approach real problems. You’ll also learn how interviewer tone and feedback can elevate – not dilute – evaluation.

More humane interviews don’t just feel better for candidates – they give your team a clearer signal, more honest conversations, and ultimately,
stronger hires.

Discover a system that monitors your team’s stress signals, staves off burnout, and evolves leadership into a dynamic and adaptive framework.

Engineering teams face challenges that traditional KPIs often overlook. In this talk, I share practical lessons learned from creating systems to detect early signs of burnout, stress, and misalignment across teams. I compare how we monitor code performance with the way we notice subtle signals in team behavior, such as variations in energy, mood, or engagement that typically remain unseen.

I will outline straightforward methods for implementing regular check-ins and feedback loops that capture early warning signs. You’ll learn simple practices that encourage transparent communication and prompt the reporting of issues before they escalate into burnout or operational disarray.

This talk is geared toward anyone who leads engineering teams, whether managing a small group or an entire department. The techniques I share are practical and straightforward, focusing on establishing routine assessments that serve as health-checks for your organization. By shifting from a reactive approach to proactive management, leaders can maintain steady progress while ensuring team well-being.

Ultimately, you’ll come away with actionable insights that help you build a self-regulating team system: one that detects early stress signals and enables timely intervention. This approach not only boosts productivity, but also helps create a sustainable, supportive work environment where innovation can thrive.

Key takeaways

• How to track subtle signals of stress and low energy in your team.
• Practical steps for setting up regular feedback and check-in processes.
• Methods for addressing issues before they escalate into burnout.
• How to balance productivity with the ongoing well-being of your team.

Discover how the Financial Times used AI to personalise paywalls and reduce churn – along with the architecture, trade-offs, and unexpected side effects of applying machine learning to real business-critical decisions.

Over the past year, the FT has partnered with two external AI providers to deliver large-scale machine learning projects focused on subscriber acquisition and retention. By using predictive models to personalise user journeys on FT.com, we’ve more than doubled our projected lifetime revenue per customer.

In this talk, I’ll walk through the real-world implementation of AI at a media organisation – what worked, what didn’t, and what we’d do differently. You’ll get a behind-the-scenes look at our architecture for integrating Sub(X) into the paywall, and how we navigated unintended consequences, like a risk to advertising revenue we initially handled the wrong way. I’ll also share how we tackled churn reduction with a different AI model built by VectorLabs, which runs on our own infrastructure – and the unique technical and operational challenges that came with that.

Finally, I’ll offer hard-earned lessons on evaluating and maintaining AI in production: how to adapt to disruptions like price changes, what to monitor during model training, and how we revisited our approach to testing to cope with the explosion of possible user journeys AI introduced. Whether you’re considering an AI initiative or already running one, this talk will give you a clear-eyed look at the trade-offs, stakeholder dynamics, and surprising complexity that come with applying machine learning in a real business context.

Key takeaways

• Strategies to minimise risk and build stakeholder confidence when integrating AI into critical user journeys
• Practical techniques for evaluating models that continuously adapt and evolve
• Why we chose different AI providers and model architectures for acquisition vs. retention – and what we learned from that decision

Promotion denial stings. Discover strategies for bouncing back, effectively managing expectations, and reigniting your career growth and fulfillment, with insights on effective communication and managerial support.

Rejection stings. Especially when it’s a promotion a person was aiming for and putting significant effort in. So what happens when that ‘next step’ doesn’t happen?

This presentation dives into the different aspects of navigating this challenging situation, offering valuable insights for both employees and managers. Understanding the emotional impact of rejection, the potential loss of trust, and the various ways disappointment can manifest are crucial first steps in effectively addressing the aftermath of a denied promotion.

This session will dive into the critical questions and decision-making process an employee faces after a promotion denial:
“Who am I without this promotion?” Having a promotion denied often feels like a personal rejection instead of a professional one. We will look into how to reframe success beyond the title.

“Is it me or the company?” It’s important to get clarity why the promotion was denied. We will examine common reasons, such as personal factors, role preparedness, or simply unfair treatment.

“Should I stay or should I go?” Should the person remain with the organization (double down on any growth areas or pursue alternative paths internally) or to seek new challenges elsewhere? We will analyze the factors that might influence this decision and provide frameworks for evaluating different options.

We will also explore the importance of well-defined promotion pathways. Transparent communication throughout the promotion process is crucial, ensuring that employees understand the criteria of performance and promotion, and have clarity on their progress. The role of managerial support in providing constructive feedback, guidance, plans, and ongoing development opportunities is key for navigating the situation successfully.

Furthermore, we will focus on the essential skill of resilience in case of a career setback. We will discover practical strategies for bouncing back from a missed promotion, effectively managing expectations and adapting to unexpected turns in their career trajectory. We will explore how to transform disappointment into renewed focus on personal and professional development, ultimately reigniting career growth and fostering a sense of fulfillment.

Key takeaways

• Techniques to transform promotion denial to an opportunity of personal and career growth
• 30-day comeback plan after promotion denial
• Mini-resilience self assessment

The definition of insanity is doing the same thing over and over again” – this quote attributed to Einstein warns us of the danger of magical thinking, hoping that trying something just one more time will achieve success when before we failed.

In this talk, I’ll argue that retrying things actually does make a lot of sense, and is in fact key to improving the resilience of a distributed system. Along the way, I’ll explain the importance of timeouts, retry limits and knowing when giving up does make sense. I’ll also show how retries can be made safe (and help avoid draining your bank account), and perhaps we’ll get to examine that Einstein quote in a bit more detail…

Closing thoughts from the host, Pat Kua.

Watch or participate in free 1:1 coaching with Mathias Meyer.

If you’re inspired by a talk, want to learn more, or just want to chat about it with other folks, head to our Office Hours area, where our speakers will be waiting for you.

Charity will be signing copies of her book ‘Observability Engineering‘.

Bring a book and get it signed – or just have a chat.