Mimi Newman · Senior Product Designer & Builder
I help B2B teams ship accessible, production-ready work faster. Seven years in product design and PM across payments, government, and travel. I design in Figma, build in Claude Code and Cursor, and bring the judgment AI can't.
Open to full-time and long-term contract roles · Los Angeles · Remote
Seven years in product design and management
Founding designer of an Apple Top 10 app
WCAG AAA on every project
Embedded client engagements of 12+ months
Selected work
Challenge. A payments processor with three audiences (partners, executives, merchants), internal tools that had outgrown their interface, and a new merchant portal to build against the leading small-business platforms.
Outcome. Three portals shipped in production code with engineering. Partner (reporting, training), executive (performance), merchant (in development).
Challenge. Locally elected officials drowning in constituent emails about potholes, zoning, and street safety, with no shared place to triage, respond, and track outcomes.
Outcome. A communication and constituent engagement platform across desktop, an agentic Outlook plug-in, and mobile, with multi-channel outreach (newsletters, SMS, press releases).
Challenge. Hopper was moving from aggregated listings to direct supply to grow margin. Engineers shipped an MVP that worked but spoke in engineer-ese, and enterprise hotel groups needed role, group, and permission granularity before they would sign.
Outcome. Redesigned core listing CRUD and reservations, built out an enterprise-grade roles-and-permissions system (8 granular permissions, 2-step wizard, plain-English confirmation), and left behind documented interaction states for the partner surface.
Challenge. Enterprise grant tools are priced for research teams ($4,999/yr Candid, $179/mo Instrumentl). Nonprofits, funders giving grantee seats, and agencies managing clients all need the same intelligence at a fraction of that price.
Outcome. Shipped end to end solo on Next.js, Supabase, and Trigger.dev: marketing site, user app, admin portal, enterprise surface, agentic ingestion pipeline (70+ tasks via OpenRouter), and end-to-end marketing automation (content, carousels, blog, Buffer, Resend lifecycle, PostHog loop). Three customer segments on one stack. $16/mo effective price.
Testimonials
“Mimi fits right into our team and works faster than we can even build.”
“Mimi really picked up what we’re trying to build and took our vision to the next level.”
“Excellent at collaborating with engineers and delivering beautiful, thoughtful UX. She takes feedback well, delivers with high output, and leads with a hypothesis-driven approach.”
Approach
AI produces screens fast. It can't tell when they don't hold together. This is the review I run on every build.
Process
Expertise
Enterprise and operator tools
At Parthenon, I designed one government CRM experience across desktop, an Outlook plug-in, and mobile. Inline editing and real undo over modals.
Dashboards and data
At Indeflex, I design the executive portal leaders use to track company and partner performance. Metric-forward, next action obvious.
Access and permissions
At Hopper, I designed the roles-groups-permissions system a non-technical hotel GM can scope without IT help.
Why me
Range
Payments, government, travel, healthcare, education. As CHANI's founding designer, I shipped v1 of an app that hit Apple's Top 10 with 10x revenue growth.
Builder
I prototype and ship React with Claude Code and Cursor, and write the steering documents that keep agents on-system. Stakeholders review working software, not static mockups.
Experiential quality
My MA in Clinical Psychology shapes how I design for how people think, decide, and feel. The layer AI can't evaluate on its own.
Systems
Built three from the ground up. Fluent inside established ones.
Design principles written for AI agents, checked into the repo, applied on every run.
Automated critique for flow, IA, copy, and accessibility, before human review.
WCAG AAA baked into every review, not added at the end.
Work lives in the repo. Engineers review a branch, not a handoff file.
About
I'm Mimi, a senior product designer with seven years in product and PM, most of it B2B: internal tools, dashboards, permissions, onboarding. Through my studio, I've embedded with teams for engagements of more than a year, shipping alongside their engineers.
This year, my building caught up with my design judgment. I designed and built Funding Findr, a programmatic grant discovery engine for nonprofits, with marketing automated end to end. I want to bring this way of working to a team where the systems I create raise the standard for everyone, not just me.
FAQ
Four questions. Does the flow hold end to end, including errors? Does the IA match how people think? Does it feel right for the moment? Does it meet WCAG AAA? AI produces screens well. It doesn't notice when they don't add up.
Daily. I ship front-end code in React and TypeScript with Claude Code and Cursor. Not a backend engineer. I build the interface, to a standard engineers can merge or extend.
Components and tokens first. If the system has a gap, I document it and propose an addition, not a one-off.
In the same repo. I deliver a branch, not a design file, so reviews focus on behavior, not translation. Worked this way embedded at Indeflex and other clients.
The leverage has shifted. One builder can ship a lot alone. The bigger impact comes from raising the standard for a whole team through design systems, steering documents, and shared prompts. That work belongs inside an organization.
Contact
Happy to walk you through how I work, live. A real problem, Figma and Claude Code, and the decisions along the way.