From C# Desktop Apps to Enterprise SaaS: 5 Years Building Production Software

Five years of building production software, starting with C# desktop applications and ending up shipping enterprise SaaS platforms, ERP/CRM systems, and AI integrations for clients across logistics, real estate, healthcare, and FinTech — here's what that path actually looked like, and what it taught me about building software that businesses actually depend on.

Where It Started: C# Desktop Applications

The earliest work was C# desktop applications — Windows Forms and WPF, tightly scoped tools solving a specific business problem for a specific client. There's a particular discipline that desktop development teaches that's easy to skip in web development: you can't paper over a slow query with a loading spinner and a "please wait" message the way a web app can, because the user is sitting right in front of the screen watching it happen. That constraint was a genuinely useful foundation for thinking about performance from day one rather than as an afterthought.

The Move to ASP.NET Core and Modern Web Development

Moving into ASP.NET Core and modern web development opened up the ability to build systems that many people, across many locations, could use simultaneously — which is a fundamentally different design problem than a desktop tool used by one person on one machine. This is where concepts like proper authentication and authorization, API design, and horizontal scaling actually started to matter in practice, not just in theory.

Two Years Leading a Team on a Real Estate Advisory Platform

The most formative stretch of this period was two years as team lead on a real estate advisory platform, leading a team of 4 developers building MLS integrations, an agent CRM, mobile apps, and listing portals used by over 200 agents. Leading a team taught a different set of skills than writing code solo does — code review as a teaching tool rather than just a gatekeeping step, breaking ambiguous requirements into concrete tasks a team can actually execute against, and the reality that a technically elegant solution nobody on the team can maintain confidently is worse than a slightly less elegant one everyone understands.

Going Freelance: Enterprise Systems Across Industries

The freelance chapter — building on Upwork and Fiverr, reaching Top Rated Plus status — is where the range of industries expanded significantly: transport management systems for logistics fleets, warehouse inventory platforms, WhatsApp-based lead CRMs with AI-drafted replies, and healthcare shift management platforms replacing paper-based rostering. Each of these looked completely different on the surface, but the underlying skill was consistent: understanding what a business actually needs (which is very often not exactly what they initially ask for), and building the smallest system that solves it correctly rather than the most impressive one that solves it eventually.

What Actually Transfers Across All of This

A few things held constant across desktop apps, team-led enterprise platforms, and solo freelance projects: relational data modeling done correctly (a lot of "the software is slow" complaints trace back to schema decisions made in week one), the discipline of treating client requirements as a starting hypothesis to validate rather than a fixed spec to execute blindly, and the fact that the hardest part of most projects is rarely the code — it's correctly understanding the actual problem before writing any of it.

What's Next

The current focus is on AI integrations layered into existing enterprise systems — not AI as a standalone novelty feature, but genuinely useful automation wired into the CRM, ERP, and platform work that's been the throughline of the last five years: AI-drafted replies in lead CRMs, document automation, and workflow automation that removes real manual work rather than just adding a chatbot for its own sake.

FAQ

If any of this sounds like the kind of experience your project needs, let's talk.