The key to driving innovation in 2026 is all about staying ahead, moving fast — and reading.
That’s according to Sasha Reiss, chief product officer and co-founder at Playground, a company that offers software that’s designed to help child care programs streamline their operations and administration. He said that the company is made up of avid readers and active knowledge-sharers, including the company’s CEO, Daniel Andrews, who publishes weekly notes that detail what he’s learned about emerging technologies and industry trends.
Reiss and his fellow leaders also encourage hands-on adoption and experimentation, providing access to tools like Claude Cowork, while recognizing and rewarding the most impactful AI-driven process improvements across the company.
“This combination of shared learning, individual investment and practical application allows us to scale quickly while staying at the forefront of new technology,” he said.
Driving innovation is about knowing when to push boundaries — and when not to. This deliberateness has enabled Vice President of Engineering Andrew Bentley and the rest of the team at Altana to refine the company’s product network, which connects businesses and governments on a common platform for global trade.
In addition to staying curious about emerging trends, his team is intentional about pursuing projects based on whether it sets the product apart from others in the industry.
“We innovate aggressively where it does and standardize where it doesn’t, and the result is a platform that lets teams move fast without the complexity tax that kills most scaling orgs,” Bentley said.
Read on to learn how Playground, Altana and nine other companies stay ahead of emerging trends while moving quickly, and how this leads to the inception of impactful products.
Featured Companies
Braze is a leading customer engagement platform that powers lasting connections between consumers and brands they love through cross-channel messaging and journey orchestration to Al-powered experimentation and optimization.
How does your team stay ahead of emerging technology trends while scaling fast?
For our high-growth decisioning services team, which is focused on BrazeAI Decisioning Studio™, staying ahead is a “push-pull” dynamic. Our AI deployment team is key to our efforts, deploying tailored contextual bandits that navigate real-world noise and latency. We act as the “exploration” engine, identifying emerging needs, like constrained optimization or multi-action selection.
“Our AI deployment team is key to our efforts, deploying tailored contextual bandits that navigate real-world noise and latency.”
We scale fast by turning these “bespoke” client wins into productized building blocks. Our product team acts as the “exploitation” engine, abstracting the deployment team’s custom logic into modular components.
Now more than ever, the competitive edge isn’t just a better model; it’s the speed at which a client-specific innovation becomes a core capability. This loop ensures we aren’t chasing trends; we are building a self-reinforcing system where every unique deployment hardens our core platform.
What recent product or feature are you most proud of — and what impact has it had?
We are most proud of our internal AI analytics framework for rapidly productizing custom client analyses for wider use. It transforms “bespoke” client-specific analyses created by our deployment teams into self-service, productized building blocks accessible to all forward-deployed data scientists and clients. This drastically reduces the time from a custom win to a core capability.
A prime example is our solution for detecting regime bias. It started as a custom analysis to identify disparities in historical regime performance. It has since been productized into a tool that uses different techniques to assess the risk of bias in our audience assignment. This proactive check is now a core part of our setup process to ensure fair experimental groups, and is now integrated into a weekly alert for continuous bias monitoring. This ability to quickly turn a client-driven problem, like the bias seen in a recent audience reshuffle, into an automated, standard governance tool demonstrates the speed and quality of our innovation loop.
How do you create a culture where innovation and experimentation are encouraged daily?
At the heart of our business strategy lies a commitment to continuous experimentation. To foster this, our product team developed a systematic testing framework, enabling us to test hypotheses and configurations to drive performance gains. We have institutionalized this approach by mandating that deployments include a test using this framework whenever feasible.
By tracking and openly sharing all outcomes — both successes and failures — we cultivate an environment where innovation is truly celebrated. The most valuable insights often come from failed experiments that reveal what doesn’t work.
An excellent illustration of this principle is a recent initiative headed by one of our forward-deployed data scientists aimed at lowering unsubscribe rates through rules-based frequency guardrails. Although that specific trial did not yield the desired results, the presentation of the findings inspired another team to pivot their strategy, ultimately leading to a successful outcome.
Another key aspect of our culture is to value individuals who identify a problem and come up with solutions and guidance. This ability to identify problems and propose solutions is at the core of one of our values, “Don’t Ignore Smoke,” which is really all about noticing issues before they turn into fires, and helping each other to find and implement a solution.
Playground’s child care management software is designed to help child care programs streamline their operations and administration.
How does your team stay ahead of emerging technology trends while scaling fast?
We stay ahead by building learning directly into how we operate. As a team, we’re avid readers and active knowledge-sharers. Dan, my co-founder and our CEO, publishes weekly notes for the team on what he’s reading, with a strong focus on emerging technologies and industry trends. Employees love to share awesome new use cases they find on social media platform X. Everyone also has a dedicated learning budget for courses, books and coaching, so continuous education is strongly encouraged.
“Everyone also has a dedicated learning budget for courses, books and coaching, so continuous education is strongly encouraged.”
At the same time, we prioritize hands-on adoption and experimentation. Every team member, regardless of role, is encouraged to build AI and coding proficiency. We provide access to tools like Claude Cowork across the company and reinforce usage through incentives. Every two weeks, we recognize and reward the most impactful AI-driven process improvements.
This combination of shared learning, individual investment and practical application allows us to scale quickly while staying at the forefront of new technology.
What recent product or feature are you most proud of — and what impact has it had?
A recent product we’re especially proud of is the Camber AI voice agent. This feature brings AI directly into customer interactions, enabling our customers to automate high-quality, real-time conversations. What makes it impactful is not just the technology itself, but how seamlessly it integrates into existing workflows and frees up admin time so customers can spend more time with their students.
Although the product launched recently, early adopters are already seeing meaningful improvements in their enrollment, responsiveness and operational efficiency. It also reflects our broader product strategy: moving beyond static tools toward intelligent, adaptive systems that actively participate in workflows. Camber is a strong example of how we’re leveraging AI to create tangible, high-impact outcomes for child care providers in service of our mission to make excellent childcare accessible to all.
How do you create a culture where innovation and experimentation are encouraged daily?
Innovation, experimentation and agency in day-to-day work are core to Playground’s culture — and that starts with us as founders. We’re a product-led company that moves fast. Our engineers ship code every day, and that mindset carries across the entire team. Whether it’s code, internal workflows or customer experience, the expectation is to ship, learn, and keep improving. Perfection isn’t the goal; progress and impact are.
When something doesn’t work, the focus is on what we learned and how we can do better next time. Dan and I actively model this by asking questions like, “What can we automate or improve to make this easier for us or our customers?” and “What can we learn from that bug? What would fix it in the future?”
We also back this up with real resources and incentives. Everyone has access to tools like Claude Cowork and Loveable, along with team-choice awards for the best “vibe-coded” tool and biweekly recognition and spot bonus for the most impactful AI-driven improvements across the team.
Altana is the world’s product network connecting businesses and governments on a common platform for global trade.
How does your team stay ahead of emerging technology trends while scaling fast?
The work here is genuinely interesting: probabilistic entity resolution at global scale, graph modeling across billions of supply chain edges, ML pipelines that balance precision and recall across massively heterogeneous data sources. The technical problems are hard, and the team gets to work on things that push the boundaries of what's possible in this domain. But we’re deliberate about where we push boundaries and where we don’t, and that deliberateness is exactly how we scale fast.
We actively track what’s emerging to understand where the landscape is heading, and our engineers are consistently evaluating new tools, frameworks and approaches in their domains. When something interesting surfaces, we run it through a clear filter: Does this create a differentiating advantage for our product? Does it make our knowledge graph more accurate, our entity resolution faster, or our data ingestion more scalable? If yes, we invest deeply, build real expertise, and stay on the frontier. If not, if it’s infrastructure that needs to be reliable but isn’t a source of competitive differentiation, we go with what’s easiest to support — battle-tested, well-understood and boring where boring is a virtue.
“We actively track what’s emerging to understand where the landscape is heading, and our engineers are consistently evaluating new tools, frameworks and approaches in their domains.”
So the short version: We’re curious and we pay attention, but we evaluate everything through the lens of whether it differentiates the product. We innovate aggressively where it does and standardize where it doesn’t, and the result is a platform that lets teams move fast without the complexity tax that kills most scaling orgs.
What recent product or feature are you most proud of — and what impact has it had?
When the Supreme Court struck down IEEPA tariffs on February 24, 2026, Altana was uniquely positioned to respond because we’d been building toward this moment for months.
The team had been at work building out our tariff stacking calculator that correctly models the interactions between six or more overlapping tariff programs. The core challenge is what we call the “30 percent problem” — the complex stacking logic where naively summing rates produces wildly inaccurate results. To give a concrete example: A shipment from Israel to the United States carries an actual duty of 28.2 percent, but if you simply add up the applicable rates, you get 143.2 percent. Getting that math right across every product, every origin and every combination of programs is a genuinely hard engineering problem. Within two weeks of the IEEPA ruling, a team of three engineers shipped a minimum viable product. That timeline would have been unthinkable even a year ago.
This wasn’t a team of 10 working around the clock. It was two to three engineers who deeply understood the domain, using agentic coding to compress weeks of boilerplate, data transformation logic and test scaffolding into days. The system they shipped ingests 100,000 raw customs entries and recalculates duties against billions of rows of tariff data. Agentic tooling handled the mechanical work so the code also got better while the product shipped faster. By early March, customers were already using the tool. Across the platform, the team had identified millions in average duty savings and calculated over one million complex product duties.
What I’m most proud of isn’t just the single feature itself. I am most proud of the system that made this possible: engineers who’d been embedded in the tariff problem for months, infrastructure that was ready when the moment came, and agentic tooling that let a tiny team ship a production system in two weeks that processes billions of rows of tariff data and puts real money back in customers’ pockets. That’s not a demo. That’s the compounding payoff of customer proximity, deliberate infrastructure investment and AI-native engineering working together.
How do you create a culture where innovation and experimentation are encouraged daily?
Give teams a clear understanding of the customer outcome they own, put them as close to that customer as possible, and then get out of the way. That's the short version. The longer version is three things working together.
First, every team needs to know what winning looks like for their customers, not in abstract key performance indicators, but in real workflow terms. When an engineer understands that a compliance analyst is spending four hours a day manually cross-referencing supplier records, they don’t need a brainstorming session to innovate. They see the problem, and they want to solve it. Customer clarity is the best innovation fuel I’ve found.
Second, we create direct feedback loops between engineering teams and the customers they serve, which is not filtered through three layers of management and a quarterly roadmap review. Engineers are hearing real reactions to what they shipped last week. That does two things: It makes people care more about the quality of what they build, and it gives them the confidence to try things because they’ll know quickly whether it worked.
Third, and this is the one most engineering orgs get wrong: You have to let teams move fast enough that failure is cheap. If every release is a six-week bet with a heavyweight review process, people play it safe. If the team can ship something in a day or two, try it with a real customer, and iterate, they’ll take smart risks constantly. Our philosophy is that getting 80 percent of things right and fixing the other 20 percent faster than competitors can ship version one is a massive advantage. But that only works if the team has real autonomy and the infrastructure to support fast iteration. Innovation happens when capable engineers understand customer context and problems deeply and can move fast enough to act on what they see.
MarketAxess’ platform is designed to make bond trading more accessible.
How does your team stay ahead of emerging technology trends while scaling fast?
We stay ahead by making experimentation part of our normal engineering cadence, not something we squeeze in on the side. We experiment freely with frontier tools and frameworks, which keeps our instincts sharp. But we go further than that: We deliberately allocate 10 to 20 percent of our capacity to fold promising discoveries into our production systems and pay down technical debt. We keep a tight feedback loop through production telemetry, post-incident learnings and developer experience metrics so we’re only adopting new tech once it can measurably improve reliability, velocity or cost.
“We deliberately allocate 10 to 20 percent of our capacity to fold promising discoveries into our production systems and pay down technical debt.”
Software engineering, like other types of engineering, relies on having sound, time-tested processes alongside choosing the right tool for the right job. You can’t know you have the right tool unless you’re continually evaluating what’s out there — and there’s no way to realistically do that unless you make the time as part of your routine work schedule.
What recent product or feature are you most proud of — and what impact has it had?
I’m most proud of the “living architecture” capability we’ve built. We use an architecture-as-code approach to generate high-fidelity technical documentation that stays constantly in sync with our codebase. The result is a real-time architecture diagram for our system, enriched with health signals from multiple data sources — observability metrics, alerting and deployment status — all in one view.
On top of that, we’re building a custom server that lets AI agents query system health directly. Anyone on the team can now debug and trace issues conversationally, without needing deep knowledge of every service and how it relates to the system as a whole. What used to take an experienced engineer precious minutes of log-diving can now be surfaced in seconds. That’s been powerful in how quickly we can analyze and then respond to production issues. More broadly, we’ve been intentional about using AI advances to improve how we actually work — not as a novelty, but as infrastructure that compounds over time.
How do you create a culture where innovation and experimentation are encouraged daily?
We deliberately carve out dedicated innovation periods where the whole team sets aside feature work and brainstorms ideas that could improve our platform or how we work. Any exciting moonshot idea gets a time-boxed spike to see if it holds up, and it’s inspiring how often it does.
We’ve already shipped features, internal tools and process improvements directly out of these sessions. The team runs casual “show and tell” demos afterward, and the energy is infectious; there’s a genuine excitement about what’s possible and where development is headed.
The key is making experimentation a scheduled, sanctioned part of how we work, not something people have to sneak in around the edges. When your team knows that exploration is valued, not just tolerated, they show up with better ideas and more willingness to take smart risks.
Findigs, Inc.’s rental screening and decisioning platform is designed to help property managers grow their communities safely and simplify the rental process.
How does your team stay ahead of emerging technology trends while scaling fast?
At Findigs, we stay ahead of the curve by pairing a modern tech stack with operating habits that help our small team scale without slowing down. We invest in reliability work that compounds over time, like tightening observability and infrastructure hygiene so we can move quickly without sacrificing stability. We also take security seriously and have integrated safe practices into our development as well as built the discipline to meet rigorous standards, like SOC 2 Type I, as our organization grows. In addition, engineering leadership gives the team room for forward-looking experimentation, including ongoing discussions about AI and how new capabilities like agentic workflows or integrating continuous autonomous development cycles can translate into real customer value rather than chasing trends for their own sake.
“We invest in reliability work that compounds over time, like tightening observability and infrastructure hygiene so we can move quickly without sacrificing stability.”
What recent product or feature are you most proud of — and what impact has it had?
One product improvement I’m especially proud of is a project I worked on as part of our broader push to modernize checkout and fee handling: moving hold-fee behavior out of a brittle, one-off configuration and into a structured model. Historically, “hold fees” are a common concept in rentals — a one-time reservation fee — but when they live in a separate configuration layer from other fees, it creates duplication, which is a confusing setup for onboarding teams and creates real risk of showing the wrong fee, or even two versions of the same fee, to renters.
By consolidating fee logic into the structured model, we can support different fee setups across properties more safely and consistently because everything is represented in one model that powers checkout, receipts and downstream workflows. The impact is fewer edge cases, less operational overhead for teams managing property setup, and a platform that’s easier to evolve as new pricing and fee structures emerge. This kind of foundational work makes everything downstream faster, from support to day-to-day iteration.
How do you create a culture where innovation and experimentation are encouraged daily?
We encourage innovation by making experimentation safe and routine; our team is expected to own work end to end and iterate quickly, but in a way that protects our customers and the platform. We have strong instrumentation so we can quickly see exactly what changed in a deployment or incident and what users experienced by using clear product and system metrics, event tracking, dashboards, distributed traces and structured logs with alerting on key service level objectives. We have careful rollout practices, but also practice blameless learning when something doesn’t go as planned.
At Findigs, high-trust collaboration is encouraged. Engineers partner closely with product and design, and we communicate frequently and share progress early so there’s room to prototype and improve the system as we build. You can see that mindset in projects like with the fee logic work, where product, design and engineering work in lockstep, pressure-testing assumptions and trusting one another to iterate quickly while keeping reliability and user impact top of mind. This allows us to rethink fundamentals like data models and flow architecture without it turning into a slow, siloed process.
Maybern’s fund management software is designed to make fund operations more efficient, accurate and seamless.
How does your team stay ahead of emerging technology trends while scaling fast?
We think about this on two levels. The first is how we work. The relationship between product, design and engineering is being reshaped by AI, and rather than just adopting new tools, which we of course are doing, we’re also forming opinions on harder questions: Where should a product manager or designer still be spending their time thinking deeply? What can they hand off to agents? That willingness to rethink the process, not just tooling, is what keeps us ahead.
The second is what we build. Staying ahead for us means putting the right primitives in place so the platform can absorb whatever comes next. Our calculation engine, MXL, is a good example: It’s a flexible, composable layer that lets us express complex fund accounting logic as building blocks rather than one-off implementations. We’re always asking: What set of primitives handles the broadest possible complexity for our users while also letting us implement and scale across hundreds of funds?
“Staying ahead for us means putting the right primitives in place so the platform can absorb whatever comes next.”
What recent product or feature are you most proud of — and what impact has it had?
We recently launched a Performance Book of Record for private funds: performance metrics powered by MXL, our calculation engine, alongside a new Report Builder that lets users evaluate and draw insights from that data. Together, they represent a meaningful shift in what Maybern is to our customers.
Our platform has been the trusted source for fund accounting calculations. We’re taking that a step further by giving fund managers the tools to draw insights from those calculations — internal rate of returns, multiples and limited partner-level returns — with the same precision and auditability they’re used to in Maybern. With Report Builder, they’re building visualizations, comparing across funds and pulling insights that used to require exporting to Excel and hours of manual work.
What I’m most proud of is the compounding effect. Performance metrics are built on MXL, the same engine that powers our core accounting, so they inherit all the rigor we’ve already invested in. And because Report Builder sits on top of both, every new calculation added becomes immediately available for analysis. Each new layer makes everything beneath it more valuable.
How do you create a culture where innovation and experimentation are encouraged daily?
Our culture is shifting as the cost of communicating ideas is being dramatically reduced. We’ve leaned heavily into rapid prototyping, not as a way to produce finished designs, but as a way to think out loud visually. We’re still writing product requirements documents, but we’re no longer waiting on a PRD to share a product idea. A working prototype or a visual walkthrough gets the conversation started faster.
Now that our product, design and engineering teams can communicate visually in addition to writing, experimentation becomes a daily habit. Someone can show you what they mean in an afternoon and get immediate feedback. When it’s easy and cheap to make your thinking tangible, people are taking more swings, and the work is feeling more creative.
Perchwell’s real estate listings platform is designed to help real estate professionals build trust with clients, collaborate seamlessly and close deals more quickly.
How does your team stay ahead of emerging technology trends while scaling fast?
At Perchwell, we try to create structured ways for engineers to explore emerging technologies without slowing down product delivery. One of the most important mechanisms we use is our internal guilds. These are cross-team, voluntary groups organized around key areas like front-end development, back-end architecture, DevOps, and generative AI. Guilds give engineers a space to discuss new technologies, share research, and evaluate tools that might improve how we build and operate our platform.
“Guilds give engineers a space to discuss new technologies, share research, and evaluate tools that might improve how we build and operate our platform.”
When a technology looks promising, engineers will often run small proof-of-concept projects to test it in a practical setting. These POCs are typically lightweight and time-boxed, and are usually driven by individuals or small teams who want to explore an idea more deeply. The goal isn’t immediate adoption, but learning, and those findings are shared broadly so the entire organization benefits.
This combination of open discussion and hands-on experimentation allows us to stay curious and forward-looking while still being thoughtful about what we adopt in production. It helps us move quickly as a company while ensuring we’re making informed technical decisions that will scale.
What recent product or feature are you most proud of — and what impact has it had?
One initiative I’m especially proud of is the work we are doing to build the next generation of the core platform that powers our products. Rather than just adding incremental features, this work is about rethinking some of the foundational systems in our architecture so we can support the next stage of both business and technical growth.
The goal is to create a platform that scales more effectively as our product evolves and as we introduce new capabilities to customers. By investing in deeper platform-level improvements now, we’re building the infrastructure that will allow teams across the organization to move faster and tackle more ambitious product ideas in the future.
What makes this work particularly exciting is the multiplier effect it creates. Improvements at the platform layer will unlock capabilities for many teams at once. That means engineers can build new features more efficiently and experiment with new ideas without constantly running into technical limitations. For a growing company like Perchwell, investing in that kind of foundation is critical to sustaining innovation over the long term.
How do you create a culture where innovation and experimentation are encouraged daily?
A big part of encouraging innovation is making experimentation feel safe and expected. Within our engineering organization, we emphasize that trying new approaches and exploring new tools is a normal part of the job, especially with the availability of AI. Engineers are encouraged to prototype ideas, test new technologies, and share what they learn.
Much of this experimentation happens organically. Guild discussions often spark ideas that engineers explore further, sometimes turning into small proposals and eventually organization-wide practices. Because our teams operate in small pods, managers and teams can decide when it makes sense to dedicate time to exploring something new.
Equally important is how we treat outcomes. Not every experiment leads to a production feature or new process, and that’s fine. What matters is that the learning is captured and shared. When engineers document what worked, what didn’t, and why, that knowledge benefits the entire organization. Over time, this builds a culture where curiosity and learning are valued. Engineers feel comfortable trying new things because even if an idea doesn’t pan out, the insights still move the team forward.
AKASA’s AI-powered platform is designed to optimize revenue cycle management for healthcare systems.
How does your team stay ahead of emerging technology trends while scaling fast?
We stay close to emerging technology by using it early and often in real work. When new capabilities appear, especially around large language models, we don’t wait for them to mature. We build small, focused experiments tied to actual healthcare workflows and evaluate them quickly. AI-assisted development tools help us stand up prototypes and test ideas in days instead of weeks, which increases how much we can learn in a short time.
“When new capabilities appear, especially around large language models, we don’t wait for them to mature.”
That speed is paired with discipline. We have evaluations that are grounded in real data and tell us if a new technique actually improved performance. We continuously evaluate performance using real encounters and user feedback, then refine from there. We work on adopting meaningful advances early while still building systems that are reliable, accurate and ready for production in healthcare settings.
What recent product or feature are you most proud of — and what impact has it had?
I’m especially proud of the progress we’ve made in helping our systems understand and reason across complex clinical documentation. Instead of looking at individual notes in isolation, we’re building systems that can interpret context across multiple documents and produce more complete, structured outputs.
The impact shows up quickly for users. Work that used to require careful review across long, fragmented records becomes faster and more consistent. This reduces cognitive load and helps teams focus their attention where it matters most. It also improves accuracy in areas that directly affect downstream processes like billing and reporting.
What makes this meaningful is that it has a much bigger impact than a simple model improvement. It’s the result of combining strong models with retrieval, validation and evaluation systems that reflect how healthcare actually works, which in turn helps ensure the output is something users can trust and rely on.
How do you create a culture where innovation and experimentation are encouraged daily?
Experimentation is part of how we work every day. Engineers are expected to bring in ideas from outside the company. We hold regular hackathons, jam sessions and design discussions, and we encourage people to build quick prototypes as a way to get buy-in. AI tools have made it easier to explore several concepts in parallel and see what works. That shift makes it much more natural to test new ideas and refine them quickly.
We also make experiments visible. Work is shared early, discussed openly and often picked up or extended by others on the team. Any result, even a failed experiment, helps us learn and refine. At the same time, there is a strong expectation around validation and quality, especially given the healthcare context. That balance allows people to move quickly, stay curious, and still feel confident in what they ship.
Granted’s AI-powered healthcare assistant is designed to help Americans navigate the healthcare system with clarity and confidence.
How does your team stay ahead of emerging technology trends while scaling fast?
Staying ahead of emerging technology is the single highest priority for our team right now. We run a dedicated “Developer Productivity” track. One engineer from each pod meets with me weekly to identify the highest-ROI investments we can make in AI tooling and infrastructure. Even our CEO, Julien, an engineer by trade, joins us some weeks because he knows just how important this moment is for our industry. The engineers in the room are closest to the actual bottlenecks, so the signal quality is high. And because it’s a standing cadence rather than a one-off initiative, we compound those investments week over week.
But engineering productivity is just the start. Every week, we challenge the entire company to find processes they could automate or reimagine with new technology. When something looks promising, we test it fast. In our most recent three-day hackathon, every team worked on ways to leverage AI to build new capabilities or create internal tooling that accelerates how we ship.
“Every week, we challenge the entire company to find processes they could automate or reimagine with new technology.”
In the next couple of years, the teams that will win are the teams that move fast and adopt new tools aggressively. We’re certain that with the operating rhythm we built, Granted will be one of those teams.
What recent product or feature are you most proud of — and what impact has it had?
We built a financial assistance feature and integrated it directly into our agent. This is a huge product win: There are millions of people who qualify for financial assistance programs and have no idea they exist. We can now surface that automatically as part of our billing case workflow, which meaningfully expands who we can help. We’ve impacted the lives of many users already, unlocking thousands of dollars of financial assistance.
The other aspect of the feature I’m equally proud of is what it proved about our system architecture. This was the first feature we developed as a module inside a newly built and highly dynamic agentic system. This matters for our velocity going forward. Instead of building each new capability into a monolithic agent, we now have a highly composable model for adding new features into the agent. The next feature we ship this way will be faster. The best product wins are the ones that also unlock a better way of building.
How do you create a culture where innovation and experimentation are encouraged daily?
Step one is to get the right people! We hire people who are comfortable without a playbook. The foundational technology behind Granted required enormous trial and error, so experimentation is how the company was built, and our core values reflect that:
“Feedback is Fuel” means we often share work at 30 percent completion, not 90 or 100 percent. This does two things. It pressure-tests ideas early, which saves time, and it makes it psychologically safe to start before you have all the answers.
“Commit to Your Craft” is about quality through iteration. We don’t need perfection on the first try. The people who improve fastest are the ones who ship, learn, and ship again.
“Play to Win” is the recognition that AI has opened massive whitespace in healthcare, and the teams that fill it will be the ones that move fast and adopt new tools aggressively. We maintain a shared channel where anyone posts new AI tools or techniques the moment they find them, and every week we challenge the whole company to find processes they could automate or reimagine. Not every experiment ships, but that’s fine. The cost of not trying is higher than the cost of a failed experiment.
Revivn offers a hardware lifecycle solution that’s designed to help IT teams effectively manage outdated electronics.
How does your team stay ahead of emerging technology trends while scaling fast?
We move fast on AI and emerging tooling, but speed without structure is just noise. The real unlock was investing early in rock-solid data architecture, getting ontology, grain and foundations right so that when new technology lands, we can actually apply it coherently and at scale. Without that, AI adoption becomes a collection of disconnected experiments that never compound.
That foundation is what liberates the team to experiment freely rather than fight the plumbing, and it’s what lets us scale new capabilities without rebuilding from scratch every time. Right now, we’re having a lot of fun in the tooling and access layer, pushing on what’s possible when the infrastructure actually supports ambition. That’s where the energy is, and it’s only possible because the ground beneath it is solid.
“Right now, we’re having a lot of fun in the tooling and access layer, pushing on what’s possible when the infrastructure actually supports ambition.”
What recent product or feature are you most proud of — and what impact has it had?
Revivn spans multiple domains, from operations and logistics to growth and donations, and each one historically came with its own definitions, incentives and data patterns. Bringing consistency across that landscape, especially at the metric and structural level, was one of the hardest and most important problems we could solve.
Through a highly collaborative build across the team, we developed a scalable, unified data model that encodes core business logic and serves as a genuine source of truth across every function. Metrics are defined once and reused everywhere. Cross-functional questions that once required navigating multiple systems and interpretations are now resolved quickly, cleanly and with the data to back them up.
The impact has been significant. We moved from fragmented, context-dependent reporting to a shared understanding of how the business actually works. But what makes it truly exciting is the leverage it creates. New tools, AI workflows and downstream applications can now plug into a stable, well-reasoned foundation instead of reinventing logic each time. The model does not just organize data; it standardizes how the company thinks about itself.
How do you create a culture where innovation and experimentation are encouraged daily?
I think it starts at the door when you walk into Revivn. There is something in the environment itself, working alongside talented people has a way of raising everyone’s game almost through osmosis. There is also an intentional level of fluid access across teams, a real openness to how ideas move through the company, and leadership that gives you the freedom, tools and confidence to move quickly and pivot when something doesn’t work.
On my team, we try to reflect that. We might start a standup by rolling dice or playing cards to decide who speaks first. More often than not, we want to hear from the more junior voices early, before the conversation gets shaped. It helps avoid anchoring bias and keeps thinking independent.
Culture is the infrastructure everything else runs on. Flattened access, earned trust and the freedom to move without friction are what make it real. Get those right, and innovation tends to take care of itself.
Kepler is developing technology to ensure the information AI platforms provide is verifiably correct.
How does your team stay ahead of emerging technology trends while scaling fast?
A lot of it comes down to the people and networks around us. We’ve built strong relationships across the engineering and startup community, and those conversations are honestly one of our best sources of signal. Talking to other founders, engineers and operators who are deep in the same problems tells you more than any newsletter will. We also give people real autonomy to explore. If someone on the team wants to try a new tool or technology, they have the budget and the room to go do it.
“If someone on the team wants to try a new tool or technology, they have the budget and the room to go do it.”
There’s also a common assumption that staying on top of new technology gets harder as you scale, but we’ve actually found the opposite. Staying ahead is one of the things that enables us to move fast. AI has been a big part of that shift. Developers with strong fundamentals can ramp up on a new technology dramatically faster now. We can prototype and pressure-test something in days that used to take weeks, so the risk of evaluating a new approach is just fundamentally lower. You’re not committing to a massive rewrite to find out if something works. You can try it, learn fast, and make a real decision. That changes the whole calculus around adoption.
What recent product or feature are you most proud of — and what impact has it had?
Our finance product has been getting incredible feedback, and the thing I keep hearing from users is that they’re using AI for work they never would have trusted it with before. That’s the outcome, but the reason it works comes down to architecture. We built what we call an agent ontology: a system where AI handles reasoning and orchestration while deterministic code handles every retrieval and calculation. Every output is traceable cell by cell back to the exact page, table and line item in the source filing. Building that required solving some genuinely hard engineering problems. We run parallel sub-agents with domain-specific tool sets and validation checkpoints between stages, so data is verified before it ever reaches the output layer. Our APIs are strongly typed so that errors function as self-correction signals for the agents. And we built a discovery layer that maps semantic differences across companies and filing types so the system can query across entities without the orchestrator needing to know every edge case. It’s the kind of infrastructure that doesn’t exist off the shelf, which is what makes it a great problem for engineers who want to build something real.
How do you create a culture where innovation and experimentation are encouraged daily?
Innovation doesn’t come from setting aside special time for it. It comes from making experimentation cheap and fast enough that it’s just part of how people work every day. We invest heavily in the tools and infrastructure that make that possible. Better build tools and better debug environments, anything that shortens the loop between having an idea and knowing whether it works. A lot of that thinking comes from our time at Palantir, where we saw firsthand how much velocity you gain when you treat the developer experience as a first-class priority. We’ve also leaned into AI across our own workflows, from how we write and review code to how we handle meeting notes and project tracking. And we hire for it. We look for people who are excited about new technologies and naturally inclined to experiment. When you combine that mindset with low-friction tooling, you don’t need to mandate innovation. It just happens.
