AutoInvent
Designing the first accessible AI patent platform.
Helping inventors file ideas and discover new ones worth pursuing.
The Problem
Patent workflows are high-stakes, unfamiliar, and easy to overwhelm with AI.
AutoInvent was helping users do several hard things at once: discover patents, reinvent ideas, draft filings, and monitor competitors. The challenge was trust and agency! Users needed strong guidance without feeling like AI was taking over their invention.
Three tensions kept surfacing:
AI suggestions could feel too final, weakening user ownership.
Core product architecture was fragmented across Dashboard, My Ideas, and filing states.
Users struggled to stay engaged when patent discovery felt too broad and unstructured.
The Solution
How might we help users navigate AI-assisted patenting without losing clarity, ownership, or momentum?
1. Unifying workflow architecture
Helped merge Dashboard and My Ideas into a clearer system that matched how users think about progress: idea, draft, ready to file, filed.
2. Making AI feel steerable
Redesigned the ReInvent flow so AI suggestions felt like starting points users could shape and stay in control over.
3. Designing discovery for retention
Explored abandoned patent browsing concepts that turned a huge content space into a more curated, repeatable daily experience.
I also worked across other core product surfaces: concepts for a competitor monitoring agent and non-provisional filing flow, plus design system components and implementation QA in Figma.
Company context
My work contributed to redesigning several of AutoInvent’s core workflows — especially AI-assisted reinvention, workflow structure, and patent discovery. During this period, the team also saw:
new weekly active users from Instagram
patent filings in one day
revenue generated that day
These are company-context signals from the same period — not metrics attributed solely to design. Design work was feeding directly into implementation and new product surfaces, including filing flows and monitoring features.
Process
Product scope
The product ecosystem I was designing across spanned invention, reinvention, filing, and monitoring:
How I worked
My process was lightweight and iterative, shaped by the pace of an early-stage startup. I focused on continuously clarifying what users needed to understand, control, and trust:
- Weekly product conversations with founders and engineers
- Competitor and pattern analysis across AI and product workflows
- 15+ user testing and feedback sessions, synthesized into recurring friction points
- Figma explorations across flows, states, and information architecture
- Design system contributions and QA as designs moved into implementation
- Concept work for adjacent surfaces like mobile and monitoring flows
Design Principles
Every principle was a tradeoff. Here's where I landed!
Users needed support without feeling like AI was taking authorship of their invention. Patent work is cognitively heavy — the interface needed to turn legal and creative ambiguity into smaller, clearer actions. And when AI makes changes, users need access to the original material and a strong sense of where they are in the workflow.
Decision 1: Making AI assistance feel steerable
In the original ReInvent flow, AI suggestions felt too much like preset answers. In a product centered on originality and ownership, that made the system feel overly authoritative.
Key changes
- Clarified that AI outputs were starting points, not final directions
- Explored bullet-point selection so users could choose what to keep
- Added a refinement step for user overrides and steering
- Moved “summarize your change” to the end of the flow
- Reframed comparison around original patent vs. final version
Why it mattered
- Made the AI feel more like a collaborator than an answer engine
- Preserved user ownership of the invention
- Kept users anchored to source material throughout the flow
Decision 2: Restructuring around user progress
Users’ work was split across Dashboard and My Ideas, making it harder to understand where something lived or what stage it was in.
Key changes
- Merged Dashboard and My Ideas into a single view
- Organized work with tags: Idea, Draft, Ready to File, Filed
- Clearer entry points by intent: ideate vs. ready to file
- Removed unnecessary progress indicators
Why it mattered
- Clarified the product’s underlying logic, not just the interface
- Matched how users already think about filing progress
- Reduced “where does this live?” friction
Decision 3: Discovery for retention
The abandoned patent feed gave users a large opportunity space, but also too much to process. The team saw retention dropping after 1–2 days, partly because the experience felt overwhelming.
What I explored:
- Daily curated patent drops by industry
- “Today” / “New” framing to make browsing feel bounded
- Category filters for quicker scanning
- Limiting visible patents per session to reduce overload
- More visual browsing — animated figure loops and action-oriented cards
A related product decision — shifting usage limits from 1,000/week to 500/day — reflected the same logic: encourage return behavior instead of one-time consumption.
This pushed the design from pure access toward sustained engagement.
Smaller exploration — AI monitoring agent
I also explored how a competitor monitoring agent could surface relevant new patents, companies, or research tied to a user’s work.
The team leaned toward contextual dashboard surfacing instead of a standalone competitor tab, since that better matched the mental model of an agent supporting ongoing work.
Key Takeaways!
AI changed what kinds of design work I could do, not just how quickly I could do it. I used AI-assisted tooling to generate variants and accelerate exploration, but the hard part remained the same: deciding how much control users should have, when AI should intervene, and how to keep people oriented in a high-stakes workflow.
This project reinforced that AI can speed up execution, but not product judgment.
Takeaways!
This project sharpened what I care most about as a designer: designing AI systems that support people without taking over; structuring ambiguous workflows into clear interaction flows; and reducing cognitive load through product logic, not just visual polish.
It also taught me that the most important AI product decisions are often about sequencing trust — what the system asks first, what it explains, when it offers help, and how it keeps the user feeling like the work is still theirs. That’s the kind of design work I want to keep doing.