Optiver is a trading firm with technology at its core. Almost everyone here is writing code as part of their day-to-day work, whether they’re quants, traders, hardware engineers, or part of our broad and diverse team of software engineers. The breadth of users and use-cases is enormous, from building low-latency trading systems to validating a trading idea via rapid prototyping with data visualisations. As the firm has grown in size, scope and technical complexity, and as AI reshapes how we work, the underlying systems and workflows that support all of this need to evolve too.

In our industry, speed matters. Data flows constantly. As market conditions shift, so do the opportunities to act. New ideas, whether from a trading desk or a dev team, are only valuable if they can be tested, refined and put into action quickly. Friction in that process costs us more than time; it undermines our ability to compete.

That mix of users, complexity and pace is why our team exists. DevEx is a strategic engineering function with a clear mandate: improve developer experience across the company in ways that accelerate delivery and reduce our time to market. We build the tools, systems and platforms that shape how development happens at Optiver. The problems we solve are high-impact and far-reaching, and they’re deeply connected to Optiver’s success.

A force multiplier in a performance-driven environment

Improvements compound; even small wins add up at scale and over time. Users maintain focus on their goal rather than struggling to make it happen.

Although we’re a relatively new team, our work touches nearly every user and system here. What excites me is how a small change can have an outsized impact. An improvement that saves seconds in a common workflow can quickly scale across more than 1,000 users. Over time, those savings compound turning seconds into minutes and hours, creating a momentum that frees up developers to spend more energy on solving meaningful problems.

Build performance shows this leverage in action. In order to reduce the amount of time developers spend waiting for builds sitting in our CI environment, we embarked on a journey of analysing job flow through our CI agents and how our build resources were being utilised. We uncovered bottlenecks that left precious capacity sitting idle, so we redesigned how jobs were assigned to agents, ensuring those resources were fully used and throughput was maximised. Build times dropped by more than 30% on average, giving developers faster feedback loops and accelerating the pace of delivery across the firm.

How our developers set up their environment is another area where small changes have created lasting impact. We saw that new joiners would spend days doing manual, error-prone tasks to set up their environment to be productive. Over the past year, we automated the workflow end-to-end, cutting onboarding time down to minutes. The benefit extends well beyond day one. Developers now get reliable, consistent setups whenever they need them, whether that’s needed for onboarding or switching projects. It’s also a platform we can build on, enabling further improvements over time.

We look across the full development lifecycle for opportunities like these: changes that let ideas flow quickly from prototype to production. At the end of the day, our success isn’t measured in trades; it’s measured in developer minutes saved, but we take pride in how that time directly enables the business to solve trading problems faster than the competition.

Our broad user base

The work we do spans the entire business. We support software engineers building our core systems, hardware engineers creating custom FPGAs, and traders and quant researchers using Python to power the research that shapes our trading strategies. Each group has its own needs and its own definition of what “productive” looks like.

A software engineer working on our trading systems wants their build to be reliable and quick. A quant researcher wants seamless access to high-performance compute clusters to develop and test models. A trader might want to open an AI tool like Cursor, prototype a quick data visualisation, and validate an idea—without having to think about virtual environments, dependencies or where their data lives. That variety makes the job more complex but also more interesting.

We collaborate directly with teams to uncover their needs, making sure our solutions are shaped around real challenges. Those challenges are varied but the aim is always the same: remove complexity where it doesn’t need to exist.

From local to global: a greenfield platform opportunity

As a firm, we began in Amsterdam, expanding organically into APAC, the US and beyond to better serve global markets. Each region built the tooling it needed to move quickly and solve local problems. That autonomy enabled a huge amount of innovation, but it also brought fragmentation: different systems, standards and duplicated effort.

As we evolve from a region-by-region model to unified global systems, we’re at a pivotal moment. We need to build a seamless platform that can support teams everywhere. It’s a greenfield opportunity to reimagine the foundations, creating a system built to scale across regions and stand the test of time. Rather than consolidating onto one of our regional CI platforms, we’re rebuilding our CI infrastructure from the ground up. Meanwhile, we’re standardising deployment pipelines so code built in one location can run anywhere without friction. This is an exciting moment: shaping the platform that will define how developers work at Optiver for years to come.

We’re designing a platform that reflects how Optiver operates today, and how it needs to scale in the years ahead.

Staying ahead with AI

As we rebuild the platform, we also have to account for AI’s growing role in the development lifecycle.

We want developers to have the freedom to experiment with new tools, which is why we’ve already invested in Cursor and GitHub Copilot to explore how AI can support writing and reviewing code. Thanks to our automated developer setup, these tools can be rolled out quickly and consistently, making adoption smooth and fast across the organisation.

AI could transform the developer experience, helping teams move faster and focus on harder problems.

Our aim isn’t to adopt every new tool on the market, but to integrate the right ones across the development cycle. That also means not just lifting them off the shelf, but tailoring AI models and workflows to Optiver’s development lifecycle to create solutions that fit our way of building software.

It’s still early days, but there’s real excitement about how AI could transform the developer experience, helping teams move faster and focus on harder problems.

Opportunities with DevEx

Improving developer experience is a challenge that spans the whole company. By embedding engineers in every region, we can support local needs while building the shared systems that will shape and power how development works across Optiver for the foreseeable future.

If you’re curious about how your skills could contribute to the challenges of our DevEx team and help create what comes next, check out our open roles.