In the trading industry, standing still is effectively the same as moving backward.
That’s according to Dr. Andrew Wendorff, a quantitative modeling lead at proprietary trading firm Belvedere Trading.
“Every system we build starts decaying in relative value the moment it ships, and we plan for that reality from day one,” he said.
To stay ahead of potential issues, the firm’s technologists focus on identifying the highest-leverage problems and solving them incrementally. This mindset, which Dr. Wendorff said lives at every level of the firm, empowers junior quants and engineers to spot small inefficiencies and fix them before they compound, a critical practice to embrace as the firm tackles the challenge of building an autonomous trading system designed to help traders more easily handle routine, repetitive tasks.
Innovation plays out on a day-to-day basis at Belvedere Trading, and at industrial equipment manufacturer Caterpillar, it manifests in largely the same way, rooted in a focus on the bigger picture.
“We start with our mission of solving customers’ toughest challenges and ask where we can make work safer, remove friction, and help people perform better,” Senior Manager of Digital Products Brad Brown said.
At Caterpillar, innovation revolves around teamwork and a customer-first mindset. According to Brown, this approach played a critical role in the development of Cat AI Assistant, which improves how the company’s customers accomplish their day-to-day work, bringing information and actions together in a single, consistent experience.
Below, Dr. Wendorff, Brown and technologists from companies such as Supernova Technology and Toro TMS describe how their team balances experimentation with stability, and how this has manifested in recent impactful projects.
Featured Companies
Caterpillar is the world’s leading manufacturer of construction and mining equipment, off-highway diesel and natural gas engines, industrial gas turbines and diesel-electric locomotives.
How does innovation show up in your company culture?
Innovation shows up less as one-time big reveals and more as something we practice every day. We start with our mission of solving customers’ toughest challenges and ask where we can make work safer, remove friction, and help people perform better.
“Innovation shows up less as one-time big reveals and more as something we practice every day.”
Additionally, our innovation stems from working together as a team. We expect teams to maintain a customer-backed mindset across product, engineering and support. That helps ensure what we build is practical and can scale.
You can see these ideals in how we approach AI. We do not think of it as a standalone feature or innovation for innovation’s sake. We see it as a way to connect trusted data, systems and decisions to make life easier for our customers so they can take confident action in the moment.
What’s one recent innovation that improved user or employee experience?
I may be biased as the product manager, but one recent innovation I am excited about is the Cat AI Assistant because it improves how our customers get work done day to day. It brings information and actions together into a single, consistent experience, rather than having customers or dealers jump between systems.
What makes Cat AI Assistant valuable is that it’s built on trusted data. It learns from feedback and preferences, making interactions more helpful and efficient. It’s also a strong example of how digital technology supports our broader enterprise goals. It helps us engage customers more effectively, showcase our technological strength, and make everyday tasks smoother across our organization.
How do you balance experimentation with stability?
We have a guiding principle at Cat Digital: to ensure that everything we develop solves a real-world challenge. So before we even get started, we’re experimenting with that goal in mind.
Once we do get started, experimentation and stability are guided by one clear priority: safety. When people rely on our machines and digital tools to do real work in real conditions, safety is not optional.
We encourage teams to try new ideas, but we do it deliberately. We start small, test in real workflows, learn quickly, and only scale what works. To our customers and to us, safety also means predictability. Whether it is a machine feature, a digital tool or an AI-driven experience, people need confidence that it will behave consistently, especially in high-pressure situations. That discipline allows us to experiment responsibly and deliver solutions people can rely on every day.
Belvedere Trading is a proprietary trading firm with offices across the United States and around the world.
How does innovation show up in your company culture?
In the trading world, competitors are always looking for an advantage, which means standing still is effectively moving backward. Every system we build starts decaying in relative value the moment it ships, and we plan for that reality from day one.
Staying competitive means identifying the highest-leverage problems and solving them incrementally. Spending months building the “perfect” solution rarely works. Markets shift, adverse selection creeps in, and every day in development is a day we’re leaving performance on the table. Innovation here isn’t about swinging for home runs. It’s about balancing solution quality against time-to-impact, knowing when good-and-shipped beats perfect-and-pending.
“Innovation here isn’t about swinging for home runs. It’s about balancing solution quality against time-to-impact, knowing when good-and-shipped beats perfect-and-pending.”
What I’m most proud of is that this mindset isn’t top-down. It lives at every level of the organization. Junior quants and engineers are constantly spotting small inefficiencies, fixing them before they compound, and thinking instinctively about how to break big problems into smaller, shippable pieces. That habit of incremental delivery, small marginal revenues, fast feedback loops and continuous improvement is the clearest expression of how innovation shows up in our culture every day.
What’s one recent innovation that improved user or employee experience?
We’re building an autonomous trading system designed to handle routine, repetitive tasks so our traders can focus on higher-level decision-making. One of the key challenges was giving our quants real visibility into how their proposed changes actually affect market behavior, something that’s inherently complex given shifting market conditions and the number of variables involved in where we choose to quote.
To solve this, the team developed a visualization tool that projects sizes across the range of potential prices at which we could quote, giving both traders and quants a clear picture of how our system is changing. For traders, it makes the system’s behavior intuitive and transparent. For quants and developers, it provides a generalized view of how their changes are performing across different market conditions.
The real innovation isn’t just the tool itself; it’s what it unlocks. Our team can now form hypotheses about system behavior before making changes, then rigorously validate or reject those hypotheses using data that directly reflects our order activity. What used to be guesswork or slow post-hoc is now a fast, data-driven feedback loop.
How do you balance experimentation with stability?
Incrementally. One of the earliest lessons I learned in trading is that you don’t need to take on the entire market to test whether an idea has merit. You can size into a position slowly, adjust prices gradually, and let the data tell you whether you’re moving in the right direction. This same philosophy applies to how we experiment with our systems. Rolling out changes incrementally reduces downside risk. I’ve learned this the hard way: What felt like a minor pricing adjustment once turned into a middle-of-the-night call to address a cascading effect that was making trading difficult.
At the same time, over-testing is a real trap. No change will ever be perfect, and you can’t anticipate every way the market or competitors will respond. Will they trade less? More? Waiting for certainty that will never come is its own form of risk. Stability without progress means quietly falling behind. The balance we strike is this: Experiment in a way that generates the data we need with the minimum disruption to live trading, and accept that every meaningful change will have some impact. Incremental delivery isn’t just a development practice for us. It’s a risk management philosophy.
Supernova Technology’s wealth management lending software is designed to automate securities-based lending from origination through the life of a loan.
How does innovation show up in your company culture?
At Supernova, innovation shows up in how intentionally we apply AI to real-world problems. We’re constantly exploring how advances in large language models and intelligent automation can simplify complex workflows for financial institutions.
There’s a strong culture of ownership, and engineers are encouraged to experiment, test new approaches, and challenge assumptions. But innovation here isn’t about using AI for the sake of it. It’s about building technology that meaningfully improves clarity, efficiency and reliability for the people who rely on it every day.
“There’s a strong culture of ownership, and engineers are encouraged to experiment, test new approaches, and challenge assumptions.”
What’s one recent innovation that improved user or employee experience?
One recent innovation was enhancing how we use LLMs to interpret and structure complex financial documents. By refining how our AI reasons through unstructured information, we significantly improved the consistency and accuracy of structured outputs. For users, that means less manual review and faster turnaround times. For our internal teams, it streamlined development and reduced time spent troubleshooting edge cases. It’s been a strong example of how thoughtful AI integration can elevate both product quality and employee experience.
How do you balance experimentation with stability?
In AI, innovation moves fast, while reliability is critical in financial services. We balance the two by building modular systems and rigorously evaluating new approaches before they reach production.
We encourage experimentation, but it’s grounded in measurable outcomes and controlled rollouts. That allows us to evolve our technology while maintaining the trust and stability our clients expect. For us, “building better tech for people” means delivering AI that’s not only advanced, but dependable.
Toro TMS offers a transportation management system for the trucking industry that’s designed to tackle the back-office challenges of bulk hauling, from load management and dispatch to accounting and driver payroll.
How does innovation show up in your company culture?
We have “Wildcard Fridays,” a day every few weeks to loosen the constraints of a typical build cycle and free up space for innovation. Maybe there’s a new technology you’ve been curious about or customer feedback that resonated, and you want to just ship it. People often deliver something genuinely useful because of our customer-first culture. It’s intended to remove friction between a good idea and put value into users’ hands. On the team side, we do visioning exercises and generate North Star artifacts early to see what resonates internally and with customers. At its best, a “crazy eight exercise” might generate multiple radically different ways to solve a problem.
“We have ‘Wildcard Fridays,’ a day every few weeks to loosen the constraints of a typical build cycle and free up space for innovation.”
We emphasize preserving high-level mental models so the product still feels purpose-built even when taking a less traditional approach. We also run a monthly AI share-out across all functions. You get fun cross-functional moments: an engineer helping build a help bot or marketing getting design help on a lead generation tool. Those collisions happen because as a builder, it feels amazing when people love what you make. It all works when leadership recognizes this time as valuable, not as time away from your “real job.”
What’s one recent innovation that improved user or employee experience?
We recently shipped Fleet Groups, a way for customers to slice their experience however they think about their trucks, whether that’s by vehicle type, dispatcher, region, etcetera. Other TMS solutions we’ve observed solve each segmentation need in isolation; one feature for this, another for that. That can feel great if it matches exactly how you operate, but is irrelevant and bloated otherwise. We try to build for flexibility instead. We’d done enough discovery to know a few different ways customers might segment, so we went in with that framing from the start.
The team combined that with a principle we call “opinionated, but flexible” — one unified experience that still feels purpose-built. We initially tested it for reporting only. But once customers got their hands on it, they showed us how it extended across the rest of their workflows. Ship a proof of concept, and users point you the rest of the way. Today, it’s defined in a single place but extendable throughout the experience. It’s the best feeling when you can take something that feels wildly heterogeneous on the surface and deliver a single unified solution that solves both today’s and tomorrow’s problems.
How do you balance experimentation with stability?
We believe you shouldn’t have to sacrifice stability to experiment. It’s about designing systems to de-risk bets while promoting intelligent failures. Concretely, we spike uncertain ideas before committing, validate at lower fidelity when possible, and assess confidence before sizing up investment. We don’t mind a longer cycle if confidence is high, but if it’s not, we find ways to learn faster. We don’t mind being wrong when it produces insight without a huge loss.
Picking the path is a messy art. There are countless decisions a team needs to make, and too much process will smother them. We default to trusting the team to make the call. That spirit of empowerment is hard to kindle and easy to extinguish if you’re not intentional about protecting it.
We have people across the spectrum. Some want to swing big, others prefer incremental, more certain steps. We see that range as a competitive advantage. Healthy tension keeps us from overindexing. The team navigates this together by disagreeing and committing, sharing assumptions, learning from shipping and talking to customers. The balance is less a formula and more a culture of continuously iterating on our approach.
Hireology offers a hiring platform for automotive dealerships, healthcare facilities and hotels.
How does innovation show up in your company culture?
Innovation at our company is less about big ideas and more about how we work every day. Teams align around measurable customer problems first, then run small experiments and problem reviews before committing to build, so progress comes from learning, not guessing. Because teams are rewarded for uncovering truth early, even when it changes the plan, improvement happens continuously rather than only after we’ve launched. We see innovation as a habit of learning faster every day, not any particular feature release.
“Because teams are rewarded for uncovering truth early, even when it changes the plan, improvement happens continuously rather than only after we’ve launched.”
What’s one recent innovation that improved user or employee experience?
Over the past year, we’ve implemented AI prototyping tools that have helped us iterate and learn at a much faster rate. It has vastly improved our ability to communicate between user experience, engineering and product on ideas, go deep on user flows to figure out what is most impactful, and very quickly put usable concepts in front of customers to validate what we’re doing. We’re building the right things faster, and it’s only getting better.
How do you balance experimentation with stability?
No one wants to feel like they’re the experiment, so we aim to make customers our partners in innovation rather than subjects of it. We test ideas early and release carefully, but the real balance comes from deeply understanding who we serve — specific industries, roles and hiring challenges — so experiments happen with the right audience in the right context. That lets us validate meaningful impact with the smallest possible change instead of broad disruption. By learning precisely before scaling broadly, we gain insight while keeping the day-to-day experience stable and trustworthy.
Quantum Rise is an AI-native company that builds a broad range of products for organizations in various industries, from private equity to consumer packaged goods.
How does innovation show up in your company culture?
We have a Consulting 2.0 mindset rooted in a products-not-projects approach. That means we don’t treat solutions as one-and-done deliverables. We keep learning loops tight, measure impact, and iterate quickly. Innovation is expected, but it’s grounded in practical delivery and measurable outcomes.
“We keep learning loops tight, measure impact, and iterate quickly.”
What’s one recent innovation that improved user or employee experience?
We introduced monthly lunch-and-learns where anyone can bring forward a topic, share a lesson learned, or demo a new approach. It has become a simple, consistent way to spread best practices, reduce rework, and continuously improve how we deliver across strategy, engineering and client delivery.
How do you balance experimentation with stability?
We balance it by starting with structured breakthrough sessions with clients to build shared AI fluency and a clear adoption roadmap before anything is built. That roadmap defines use cases, priorities, success metrics and operating requirements so experimentation stays focused and what moves forward is designed to be stable, measurable and ready for real operations.
Caxy builds custom software solutions for organizations across a wide range of industries, including financial services, healthcare and hospitality.
How does innovation show up in your company culture?
I think one of the nice things about being a smaller company without as many regulations and slow processes is that we’re able to experiment a lot quicker. As an agency, we work on many different types of projects, which gives us a really good advantage, as we can try different approaches across different contexts. With AI in particular, we started by playing around with smaller concepts. But once we realized it could be pretty good for our developers and speed us up, we then started getting Microsoft Copilot for people and introducing them to how to use it. We did a lot of experiments on the side.
“I think one of the nice things about being a smaller company without as many regulations and slow processes is that we’re able to experiment a lot quicker.”
Once we felt good with a baseline of how we work with it, we decided to try it on a new project fresh. Being able to test it on something new without legacy constraints was really helpful. We could see what worked without things going off the rails. We’re able to rein things in quickly and keep an eye on it from a higher level.
Mike LaVista, CEO: We encourage people to try things. If they go a little further than we expected, that’s actually good. We tested the edges and brought it back. Being a smaller company, we can catch things before they go too far.
What’s one recent innovation that improved user or employee experience?
Sticking on the AI path: We definitely have developers here who were hesitant and skeptical about AI. On a recent client project, there was a lot of work done to set up Copilot to be really useful for particular things in the codebase.
By doing that setup work, it actually helped with the skepticism. Once you use it, and it finds something you didn’t think of or catches something in code review that you would have missed — something that’s kind of a bigger thing — you realize there’s definitely power to this. And it didn’t code everything for you; it just reviewed your work. Showing small glimpses of proof that it can be effective and powerful has been really good for getting the team bought in.
LaVista: Adding AI into the review process has been interesting. Having automated code reviews on everything has helped us catch more things. It’s also driven more communication on pull requests because often AI will recommend something that’s not necessary or is flat-out wrong. That requires you to think through it and respond. It’s created more conversation, not less.
How do you balance experimentation with stability?
This is a big one. For us, it’s about introducing innovation with a lot of guardrails. It’s “human plus” — we’re still doing all of our human checks. AI is essentially an additional check on top of what we already do with our process.
That way, we don’t just let it run and suddenly have big issues emerge. We introduce innovation in a way that isn’t totally disruptive. We run it alongside our existing process to understand where it fails and where it doesn’t. That’s where experiments have been helpful.
LaVista: I’d add another dimension: How do you balance experimentation with value? You could go down endless rabbit holes trying things. At a certain point, you have to make a judgment call: Is this experiment worth it?
We try to look ahead. If this worked at the end of the tunnel, would it be somewhere we wanted to be? If so, it’s worth pursuing. But sometimes you’re experimenting with something that doesn’t really have the biggest outcome. Knowing when to stop is as important as knowing when to start.
