Jason Sonderman · Product design and design leadership · 15 years leading B2B SaaS UX

Bring me the problems that don’t neatly fit inside a design ticket.

The hardest product problems rarely belong to design alone. They cross product strategy, research, technology, systems, teams, and the way an organization builds software.

I get into the work myself, from understanding the problem to making ideas testable. Then I help turn what we learn into product direction, systems, and better ways of building.

Complex product · Arcos Control Tower

Jason’s testable Control Tower command center prototype, showing a restoration map and crew list. — **From prototype to product.** I created the testable concept and led the design delivery that followed. Senior designers developed the near-final screens.

**From prototype to product.** I created the testable concept and led the design delivery that followed. Senior designers developed the near-final screens.

The product model changed when we followed the work.

Field research showed that updating legacy interfaces would miss the sequence of storm restoration work. That shifted the direction for Arcos Control Tower and Arcos Field toward a product model built around the work itself.

In production: designs for Mobilize destinations and Convoys, plus sign-in, in the first MVP release, August 2025. User profile management and rosters followed in July 2026. In beta: Command Center with desktop timecards, September 2026.

Reusable system

The products needed a system behind them.

I built Harmony to carry design intent through tokens, component guidance, and implementation context. The aim was to make recurring decisions usable by designers, engineers, and AI coding agents as the products were built.

Explore Harmony

Harmony’s Chip documentation. A table maps each chip color to an intent: default is neutral, for generic tags and filters; primary is the main category or active selection; secondary is a sub-category; success means completed, active, or approved; error means failed, rejected, or blocked; info means in progress; warning means pending, review required. Below it, the same four work order status chips, Open, In Review, Closed, and Overdue, are shown in the light theme and the dark theme. — **One source, three audiences.** Harmony’s chip guidance maps each color to an intent, so “warning” means “pending, review required” in light and dark alike. Designers, engineers, and AI coding agents read the same guidance, agents through Harmony’s Storybook MCP.

Way of building

Make the idea real sooner.

A hypothesis becomes more useful when people can try it. The first Timecards prototype I coded for Arcos Field had each person entering their own time. On the floor of the Linemen’s Rodeo, linemen and foremen showed us that in a storm, the crew lead enters time for the whole crew. The prototype changed to match, and utility leaders tested the revision. Caught on a convention floor, that was a prototype revision. Caught in build, it would have been a schema change.

See what the Rodeo changed

Early Timecards prototype, Schedule Verification screen. It asks “Did you work your scheduled hours?” for one person, with a 06:00 to 22:00 schedule and buttons to submit or edit the hours. — **Make it testable.** The same step, before and after the Linemen’s Rodeo. Linemen tried the left version on the floor. The right version is the revision utility leaders tested next, before any production code was written.

Revised Timecards prototype, the same screen. It asks “Did everyone work their scheduled hours?” and lists four crew members under the same 06:00 to 22:00 schedule. — **Make it testable.** The same step, before and after the Linemen’s Rodeo. Linemen tried the left version on the floor. The right version is the revision utility leaders tested next, before any production code was written.

Explore selected work

A question worth asking

What’s actually limiting the product?

More often it’s the workflow, the decisions behind it, what we’re learning from customers, or what gets lost between design and code.

I’d start by finding out, and let the evidence decide what changes.

Solve the problem. Create leverage from the solution.

The artifact isn’t always the end of the work. What I learn can change the product, the system behind it, or the team’s ability to solve the next problem.

When I say shipped, I mean live in a customer release. When I mean delivered to engineering, I say handed off.

Research into shipped mobile work

Arcos Field

I led the field research and design delivery that shaped Arcos Field around the sequence of storm restoration work. The first MVP shipped in August 2025 after two scope reductions, giving field crews a first step into the new platform.

In production: convoy creation, the event picker, profile manager, and sign-in in the MVP, August 2025. Timecards and the AI Expense Agent followed in September 2026.

Read the Arcos Field story

Arcos Field crew timecard showing daily status, shift start and end times, crew members, notes, total hours, and options to send or save a draft. — **The work in the field.** Crew time reporting and convoy travel preparation, shown through captures of the recreated interface. My role was research and design leadership; the design team developed the final screens.

Arcos Field convoy screen showing an editable arrival estimate, travel status, destination, support, and the Start traveling action. — **The work in the field.** Crew time reporting and convoy travel preparation, shown through captures of the recreated interface. My role was research and design leadership; the design team developed the final screens.

Design sprints and a reusable system

BetterCloud Secure

I led research, design sprints, and prototyping to help IT teams act on security risks. Each sprint focused on a new problem and ended with user testing that week. Fulcrum supplied the shared components behind the screens, so the system supported the work as the product took shape.

Handed off, partly in beta, never in full production. The secure file grid with inline actions, bulk actions, grouped file-category tabs, and first file-count insights reached beta. A backend caching issue, keeping data current with customers’ systems, held up the full release before the project was shelved. The Secure home with deeper insights was handed off but never built.

Read the Secure story Explore Fulcrum

BetterCloud Secure risk inbox showing risk lifecycle tabs, sensitive-content insights, filters, and a list of publicly shared files. — **From finding risk to acting on it.** A reconstruction of the Secure designs using Fulcrum components. Parts of this reached beta. The full release was held up by a backend caching issue before the project was shelved, and the design system continued to serve the platform.

Organizational capability

TVH Uplift

At TVH, I made the case to move design system engineers under UX, then co-designed governance with an engineering advocate. Uplift grew from one team using it unguided to four brand tech teams.

In production use: four brand tech teams, 15 front-end developers building on the system.

Read the Uplift story

Looking for a specific path?

Beyond these projects

The person behind the work.

I’ve spent much of my career building with teams and helping designers grow. There’s more about my experience and what matters to me on the pages below.

Explore experience More about me

Jason smiling in a café, wearing glasses and a burgundy shirt.

Next conversation

Tell me what you’re working through.

I’d like to understand the problem before talking about a solution.

Email Jason

This site works the same way. Each case study has a machine-readable twin generated from the same source as the page. How it’s built