“Turn your holiday into a construction-site showdown…”

That was the opening line in the concept we shared with the Huseman Group when they asked us to create something memorable for their clients. It needed to reflect their market-leading construction services, feel thoughtful, and incorporate AI in a meaningful way.

The idea was ambitious: a strategic construction-themed card game with custom artwork and real gameplay mechanics. Players would complete building projects by assembling crews and collecting materials. A physical deck would anchor the experience, while an AI companion would generate assignments, balance gameplay, and make every round unique.

Sounds cool. We just had to make it work.

Design First. Then AI.

At its core, Heroes of the Build follows a familiar rhythm: draw, discard, and assemble combinations, similar to rummy. The 52-card deck represents building materials and team members such as lumber, brick, steel, carpenters, project managers, and craftsmen. Players compete to complete construction projects by laying down specific three-card combinations.

Each player receives a unique construction project generated by the AI. One might be building a boutique hotel. Another might be tackling a distribution center. Each project includes multiple three-card tasks that shift based on player count and overall resource distribution.

If too many players need the same materials at once, the game stalls. If one project is easier than another, competitiveness fades. The AI evaluates player count, available materials, and potential overlap to generate projects that feel different but remain equally achievable. It maintains balance so that winning depends on decisions rather than an unlucky assignment.

Sharpies Before Software

Before the AI ever generated a project, there were handwritten cards spread across a table.

We built the first prototype by hand. We scribbled material values, mocked up project cards, and played full rounds. Then we broke it. Certain cards flooded the table. Some combinations dominated. Other turns dragged.

So we adjusted. We redistributed materials. Refined the three-card constraint. Tightened win conditions. Then we played again. It broke again. But we kept going.

That analog phase mattered. At Ample, we believe technology works best when it supports a clearly defined human experience. AI is excellent at managing variables, but it needs boundaries. We had to understand the system before asking it to optimize anything.

The conversations around the table were simple: Is this fun? Is it too complicated? What happens if we change a rule? Those questions built the framework the AI would later operate inside.

Stress-Testing the System

Once the physical version felt solid, we built a digital testing environment. This allowed us to simulate gameplay quickly and uncover issues that might not appear during a handful of in-person sessions.

We could see how small adjustments affected the entire system. That loop of human intuition paired with digital simulation strengthened the game. It also gave us confidence to send the physical cards to print, knowing they had been battle-tested and that the rules and components were sound.

Bridging Analog and Algorithm

Each deck includes a QR code that connects players to a companion site where they meet their AI foremen, learn the rules, and start a game. Unique player codes allow others to join, and each participant receives a different objective within the same physical game. Their decisions, along with a bit of luck, determine whether they save or sink the job.

The cards create the tactile experience around the table. The AI companion adds dynamic assignments and evolving scenarios. Together, they connect physical and digital in a way that feels intentional.

What We Took Away

AI works best inside structure. It balances constraints, generates variation within defined rules, and maintains fairness in complex systems. It supports creativity and strategy rather than replacing them.

In the end, we built a game people genuinely enjoy playing. The AI works quietly in the background, keeping things dynamic and fair, while the real engagement happens around the table.

That’s the kind of AI we’re interested in. Practical. Purposeful. Built to strengthen good ideas rather than overshadow them.

Interested in moving to the JAMstack? Let's talk.

Want to stay in the know on what we know?

Sign up for our email newsletter. Nothing spammy about it. Just a monthly rundown of what we’re sharing.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.