MNTR

MNTR helps experts monetize their knowledge through mentorship, advice, and feedback — an AI mentorship and strategy platform for storytellers and founders, with chat, playbooks, courses, and missions. My engagement focused on strengthening a founder-built application that already had users and revenue.

MNTR began as a Replit prototype with users. It needed multi-tenancy, role-based access, security, and separation of data and AI context. The founder also wanted to retain the speed of building features directly in Replit.

The goal: preserve founder-led experimentation while giving the growing application a stronger architecture and release process.

I served as Solutions Architect and AI Solutions Architect, working directly with the owner on business needs, the solution plan, and priorities. I structured the work with my engineers and kept the owner involved through regular progress updates and delivered outputs.

I shaped a working model in which the founder continued building features and Assembled Systems provided engineering review and production preparation.

The founder builds and experiments in Replit, keeping the speed needed to put ideas in front of early users. Changes then pass through engineering review: pull requests are managed, release issues are cleaned up, and architectural concerns are addressed. Assembled Systems prepares and deploys the production update, and the tenant and access boundaries keep users and their context separate.

Goal-oriented onboarding, a growth audit, and a restrained green accent. The app prompts users to explain what they are trying to build, with an option to defer, so guidance can begin with useful personal context.

An Express and PostgreSQL backend with Drizzle. Gemini and Perplexity support different reasoning and retrieval tasks; Stripe supports credit-based monetization and Twilio off-platform communication. Our confirmed additions are multi-tenancy, role-based access, and context separation.

The plan covered a complete audit, Replit cleanup, new onboarding, and scalability work, with a first milestone target of moving 50 users. The lasting engagement model is founder autonomy combined with engineering governance — a team responsible for preparing releases and publishing them.

The tenancy, access-control, and context-separation capabilities are implemented in the application. The founder continues creating features in Replit, with an engineering team reviewing and preparing the changes for production.