← Back to projects
Mektra preview

Mektra

A software studio building custom web applications, CRM systems, and revenue infrastructure for growing businesses.

2025 shipped Founder & Lead Developer
  • Next.js
  • React
  • Node.js
  • PostgreSQL
  • Prisma
  • Vercel
  • Stripe
  • Tailwind CSS
  • TypeScript

Tech Stack

Next.jsReactNode.jsPostgreSQLPrismaVercelStripeTailwind CSSTypeScript
mektra.config.ts

Problem

Small and midsize businesses are stuck between enterprise software that doesn’t fit their workflows and generic SaaS tools that force them to adapt. Most can’t afford a full dev team, and agencies tend to over-promise and under-deliver. The result is companies running critical operations on spreadsheets and duct-taped integrations.

I saw this firsthand across the mortgage, real estate, and sales industries. Businesses paying $50K+ for tools they use 20% of, while the workflows that actually drive revenue run on manual processes.

Why this solution

The agency model is broken for most SMBs. You get a salesperson, a project manager, a designer, and a junior dev. Four people, none of whom understand your business. I built Mektra to be the opposite: one person who understands both the business problem and the code, working directly with the client from discovery through deployment.

Constraints

  • Solo founder managing all development, sales, and operations
  • Clients need production-ready software, not prototypes
  • Must be profitable from project one. No runway, no investors
  • Every project has unique business logic. No cookie-cutter templates

Tech Stack

  • Next.js + React for the frontend. Server-side rendering for performance, React for interactive UIs. The App Router gives me clean API routes alongside the frontend without spinning up a separate backend.
  • Node.js powers the API layer and server-side logic. Keeping frontend and backend in one language reduces context-switching and makes it easier to maintain as a solo developer.
  • PostgreSQL for the data layer. Relational data modeling is the right fit for business applications where data integrity matters more than write speed.
  • Prisma as the ORM. Type-safe database queries that catch schema mismatches at build time, not in production. Migrations are versioned and reproducible.
  • Stripe for payments and billing. Handles subscriptions, invoicing, and webhooks so I don’t have to build payment infrastructure from scratch.
  • Vercel for hosting and CI/CD. Push to main, it deploys. Preview branches for every PR. No DevOps overhead for a solo operation.
  • Tailwind CSS for styling. Utility-first approach keeps styles colocated with components and eliminates dead CSS across projects.
  • TypeScript everywhere. End-to-end type safety from database schema to API response to component props. Catches entire categories of bugs before they ship.

Key decisions

  • Founder-led model over agency scaling. Quality and trust over volume. Every client gets direct access to the builder.
  • Full-stack ownership. I handle everything from data modeling to deployment, which eliminates the coordination overhead that kills most agency projects.
  • Weekly demo cadence. Clients see real progress every week. No month-long blackboxes.
  • Revenue infrastructure as a service category. Most studios build websites. I build the dashboards, automations, and CRM systems that actually drive revenue.
  • Boring stack, interesting problems. The tech choices are deliberately stable and well-documented. The creative energy goes into solving the client’s business problem, not wrestling with bleeding-edge tools.

Outcome

Mektra is live and serving clients across multiple industries. The studio model proves that deep technical ability combined with business understanding creates better outcomes than traditional agency structures. Clients get working software faster because there’s no telephone game between stakeholders and the person writing the code.

Lessons learned

  • Discovery is the most valuable phase. The projects that go smoothest are the ones where I spend the most time understanding the business before writing any code. Rushing to build costs more time than it saves.
  • Scope is the enemy. Every client wants to build everything at once. The best outcomes come from shipping a focused v1 and iterating. This is harder to sell than it sounds.
  • Solo doesn’t mean isolated. I rely heavily on well-documented tools and active open-source communities. The advantage of a solo model isn’t independence; it’s eliminating unnecessary coordination.