Embedded product design
I join your standups and backlog, get up to speed on your domain in weeks, and make the product decisions where the trade-offs happen.
Deepest track record: Energy software.
1 year project
2 years project
1 year project
Trusted by:
Design gets looped in after the architecture is already set.
Your engineers make product decisions with nobody in the room to push back.
The last designer needed three months to understand the domain, then left before it paid off.
Your standups, your backlog, your Figma, your tickets. No design silo, no handoff queue, no deliverables pack that someone has to translate before it can be built.
I can read the ticket, hold the argument about the trade-off, and tell you roughly what a decision costs to build. Most designers can do one of those three.
Grids, devices, tariffs, regulation, telemetry. In a new domain I'm useful in weeks rather than quarters, because getting up to speed fast is the job.
Days 1–5
Access, rituals, domain immersion
Into your tools and your standup. Reading your docs and asking your engineers the naive domain questions early, while it's still cheap to be wrong.
Days 6–15
First concepts, deliberately rough
Low-fidelity on purpose, so you can tell me where I've misread the domain before anything is polished enough to feel expensive to change.
Sprint 1
Owning real decisions
Work goes into your tickets and your release, not into a separate deliverables folder. Decisions get made with the people who have to build them.
Ongoing
Month two is faster than month one
Because by then I'm no longer asking what a settlement period is. The compounding domain context is the whole point of embedding rather than contracting a deliverable.
Engagements
EV-charging platform · distributed engineering team · long-term embed
01
The client moved fast, and during the workshops the Figma file, our source of truth, changed every day. The dev team fell into a rabbit hole and was soon blocked.
02
I added Zeplin to the stack. Once the team and client agreed, I moved the approved files into Zeplin as our new source of truth.
03
The dev team was unblocked and could focus on what was approved, while the product team kept discovery and planning moving in parallel.
04
Each role stopped looping the same ground. Our PM and the client's stakeholders started pulling in the same direction.
Power2Drive
Europe 2025 · Munich
" One senior designer inside your team. Not a rotating pod, not a junior on the account.
I've spent my career inside engineering teams rather than adjacent to them, and I've managed distributed dev teams directly. That's why I can hold the trade-off conversation with your engineers instead of waiting for a spec to arrive.
I started HUKI because I believe the person on the other side of the screen deserves an interface they can use well, no matter how complicated the system behind it is.
What clients raise unprompted: engagement, the ability to run a dev team, and clear communication. That's the job as much as the design work is.
The core engagement
2 days a week · scales to 5
I'm on your team, in your rituals, owning product decisions end to end. Consistent days each week, agreed up front, so nobody is competing for my calendar.
Approach
3-month minimum · 30 days' notice either side · invoiced monthly in advance
Start here
3-day minimum, prepaid
I join for a few days and work a real problem in your backlog. The lowest-risk way to find out whether this works before anyone commits to a month.
Scoped on the fit call · no minimum term · rolls into a retainer if it works
Bounded scope
Scoped on the call
2 weeks of domain immersion and product discovery, or an audit of an existing product. Useful, but it isn't the main thing I do.
Fixed scope and date agreed up front · half the fee credits a retainer
Depends on the engagement. The fractional embed runs as a monthly retainer scoped to your days per week, from 2 up to 5; the trial sprint is a small prepaid fixed fee built to de-risk the decision; the discovery sprint is fixed-scope, with half the fee crediting a retainer if we continue. I'll give you a real number on the fit call once I know the shape of the work.
I plan around it like any senior hire would: decisions and context live in your tickets and docs as we go, not just in my head, so a short absence doesn't stall the sprint. For anything longer, I flag it early and we adjust the week together rather than you finding out from a missed standup.
The fractional embed starts at two consistent days a week and scales to five as the work grows, agreed up front so nobody's competing for my calendar. I work Kraków / CET hours and keep enough overlap with distributed teams to be useful live in standup, not just async in the backlog.
Either side can end the retainer with 30 days' notice once the three-month minimum is up. No long tail, no surprise auto-renewal. If we're mid-decision on something, I'd rather hand it off cleanly than walk away at the invoice date.
It doesn't. I work inside your existing system rather than replacing it. Most engagements start by inheriting your Figma library and your patterns. If there's already a designer on the team, I plug into the gap they don't have time for, not into their seat.
Energy is my deepest track record, not the only one. I've run the same embedded model in EV charging and smart-charging control layers. What matters more than the domain is a complex product with real trade-offs; the ramp-up approach is the same regardless of the vertical.
Access to your tools, like Figma, the backlog and the standup, plus one person on your side who can answer the naive domain questions in week one. Nothing else. I don't need a brief; I need to be in the room where the decisions happen.
You do. Everything I produce, including files, decisions and documentation, is yours outright, with no license-back or retained-rights clause. That's standard in every engagement, not an upsell.
Yes, both are standard practice. I'll sign your NDA before any domain specifics are discussed, and every engagement runs under an MSA covering IP ownership, confidentiality and notice terms. Happy to work from yours or provide mine.
Give it 30 minutes.
A short call: what you're building, what makes it complex, and where product decisions currently break down. No deck. If I'm not the right person for it, I'll tell you on the call.
Not ready for a call? The one-page overview (PDF) No email required.
A couple of lines on what you're building. I'll reply to scope the trial.
A couple of lines on what you need scoped, whether that's discovery, an audit, or something else, and I'll reply with next steps.
