Launching a Multi-Agent AI Workflow Platform from Zero to Paying Customers
SketricGen is a canvas-based platform for building and orchestrating multi-agent AI systems, built on OpenAI's Agents SDK. As part of the founding product team, I owned end-to-end product delivery from concept to a live product with paying customers.
Role
Technical Product Manager
Company
Sketric Solutions
Team
4 to 5 people
Stack
AWS, OpenAI Agents SDK, Stripe, Figma
Challenge
Building an AI orchestration platform from zero requires solving for both technical complexity and user experience simultaneously. Multi-agent systems are inherently complex to configure, and the target users, technical teams and developers, have a low tolerance for onboarding friction. The challenge was to build a platform powerful enough to orchestrate real agentic workflows while keeping the path to a first working agent under five minutes.
Results
SketricGen reached paying customers within one build cycle, with 700+ signups at launch. The redesigned onboarding flow, built around user background to personalise the experience, cut the time from signup to first completed agent build from over 8 minutes to under 5 minutes, directly improving activation rates at the top of the funnel.
5 min
First agent build time
700+
Signups at launch
1 cycle
Time to paying customers
Process
Product Philosophy and UX Design
Defined the product philosophy and long-term vision for the platform. Designed user journeys and flows in Figma, focusing on reducing the cognitive load of multi-agent configuration through a visual canvas interface. Designed personalised onboarding by user background to minimise time-to-first-value.
Infrastructure and Build
Owned backend infrastructure built on AWS using Cognito for authentication, DynamoDB for data storage, and Lambda for serverless compute. Integrated OpenAI's agent orchestration frameworks and managed payment gateways and subscription lifecycle via Stripe.
Sprint Execution and Roadmap
Led sprint planning and ticket management using Linear. Maintained product roadmap and documentation in Notion. Led QA sessions and feature validation. Managed cross-functional handovers and release readiness across a team of 4 to 5.
Growth and Acquisition
As the product stabilised, transitioned into a growth-focused role defining acquisition strategies, SEO and AI-driven content initiatives, and paid user acquisition channels.
Stack
Conclusion
SketricGen represents the most complete expression of end-to-end product ownership in my career to date. Taking a technically complex AI product from zero infrastructure to a live platform with paying customers, while simultaneously managing backend architecture, payments, UX design, and growth strategy, is the clearest demonstration of what a Technical PM with a growth and analytics background can do when given full ownership of a product from day one.
