OPPRO — AI ASSISTED PRODUCT DESIGN
Organizing an AI-powered agency to connect strategy, design, and product creation.
At Oppro, I define a working model that combines human direction with a coordinator and 12 specialized agents in Claude Code. I organize roles, instructions, and context to tackle strategy, design, development, and business tasks.
Through this process, I built three proprietary tools: a UX/UI auditor, a foundations and tokens generator, and a commercial pipeline.
Role
AI-Assisted Product Designer
Company
Oppro
Year
2026
platforms
Claude Code
My role
As a Product Design Lead with hands-on experience and intermediate knowledge in AI-assisted design, I designed and structured a one-person agency operating model based on specialized agents within Claude Code.
I organized the functions of strategy, product design, marketing, sales, development, operations, and data analytics using Markdown files with distinct instructions and responsibilities.
I also defined an overarching coordination structure and project documentation, organizing agentic workflows while maintaining a human-in-the-loop approach where strategy, priorities, design decisions, and review of deliverables remain under human oversight.
This project explores how to apply Design Operations and team organization principles to new, AI-assisted work models.
The challenge
Design an operational model for a one-person agency capable of tackling digital projects that require different specialties: strategy, UX/UI, Design Systems, development, and business.
The challenge consisted of organizing artificial intelligence capabilities into a coordinated system, avoiding working with isolated agents, inconsistent instructions, and processes that are difficult to maintain.
To solve it, it was necessary to define responsibilities, structure the shared context, establish coordination mechanisms, and maintain human supervision over strategic decisions and the quality of the results.
Simpler management, a clearer experience
I designed a structure with central coordination and specialized agents, guided by my product and design judgment.
I applied this model to the construction of three web tools: a UX/UI auditor, a foundations and tokens generator, and a commercial pipeline.
01
Strategy and scope
I defined the agency's focus and the tools I wanted to build, connecting business needs with design and automation opportunities.
02
Agent Organization
I distributed responsibilities between a coordinator and 12 specialized agents to handle different tasks in the process.
03
Instructions and context
I designed prompts, instructions, and roles, using shared context in Markdown files, to guide agent workflows in Claude Code.
04
Design and construction
I led the definition of flows and interfaces as well as AI-assisted building, fine-tuning the tools throughout development.
Leadership responsibilities
Define the vision and priorities of Oppro.
Organize roles and responsibilities of the agents.
Establish instructions and work context.
Lead product, UX/UI, and design system decisions.
Coordinate the construction of in-house tools with Claude Code.
01
Specialized coordination
A coordinator and 12 agents with defined roles and instructions.
02
Three web tools
A UX/UI auditor, a foundation and token generator, and a sales pipeline.
04
Design and construction
I led the definition of flows and interfaces as well as AI-assisted construction, fine-tuning the tools throughout their development.
03
A basis for evaluation
I structured Markdown files to define the purpose and instructions for each role and to provide context for working in Claude Code.
Impact
1
Management and coordination
A framework connecting strategy, design, development, and business through human leadership and AI assistance.
Three of Clubs
Specialized coordination
A coordinator and 12 agents with defined responsibilities to organize the work in Claude Code.
☺︎
Instructions and context
Markdown files that define each agent's purpose and guide their work toward the project's goals.
❖
Three web tools
A UX/UI auditor, a foundations and tokens generator, and a commercial pipeline, built using AI-Assisted Design.
Proposed Evolution: Agent-Ready Design Systems
The next proposed step connects this practice with my experience in Design Systems: structuring tokens with explicit values, components with variants and states, and usage rules so that agents can consult the relevant context for each task.
The evolution toward an agentic system would incorporate change proposals subject to human review and defined autonomy limits. This capability is not yet implemented or validated in Oppro.
Learning
Organizing work with AI requires making decisions explicit that often remain implicit within a team: who does what, with what context, and under what direction.
Building Oppro allowed me to explore that organization and apply it to concrete products. The next step is to evaluate its usefulness, detect errors, and adjust the process based on evidence.
The tools do not yet have user validation or a formal QA process; the results of use and impact remain to be measured.