Written by Tuan Nguyen · Published · Updated
Most business websites sit there looking finished and do almost nothing. They load a logo, a nice photo, a few lines about the company, and then they stop. No lead captured, no question answered, no busywork taken off your plate. It looks like an asset. It behaves like a brochure you paid a lot for and pinned to a wall.
I am Tuan Nguyen, a full stack developer in Houston. I build websites and web apps for business owners who need the thing to actually do a job. Book the call. Filter the lead. Answer the question that gets asked forty times a week. Take the payment. Run one slice of the business while you are busy running the rest. I use AI every day to build faster, but I do not hand it the judgment that decides whether what you paid for is any good. That part stays with me.
What I build
Most of my work lands in one of a few places.
Business websites that actually convert, built around one job instead of ten decorations, so a visitor turns into a lead or a customer instead of quietly leaving without a trace.
Custom web apps, the dashboards and booking systems and client portals and internal tools that finally replace the spreadsheet your team has been fighting for two years. If you are not sure whether you need an app or just a website, that is worth settling first.
Automation, the quiet workflows that handle intake, follow ups, data entry, and routing, so nobody spends an afternoon copying things between tabs.
Integrations, connecting your site to the CRM, the payment processor, the calendar, whatever you already run on.
And first versions of product ideas, the SaaS MVP that takes a sketch and turns it into something real people can log into and use.
If it touches the web, data, users, forms, payments, or workflows, there is a good chance I can help shape it into something useful.
Where AI fits, and where it does not
I want to be straight about this, because a lot of people are selling the opposite. AI is a real multiplier when I write software. It scaffolds, it drafts tests, it catches things, it explores options in minutes that used to take hours. I would be slower and worse without it.
But it is only useful when the person driving already knows what good looks like. AI does not decide how your data should be structured. It does not own your security when something breaks at 2am. It does not understand what your customer needs on a Tuesday afternoon when they are comparing you against three other options and about to choose. That is judgment, and judgment is the thing you are actually paying a developer for. So the rule is simple. AI does the speed. I stay on the hook for the result.
The part most builds skip
The parts that show well in a demo are easy. Animations, a trendy layout, a hero image that moves. The parts that decide whether the thing works are boring and invisible. Real authentication. A database that still makes sense a year from now. Pages that stay fast when the traffic is real. Security that holds. A clean path from a stranger landing on the page to you getting the lead.
Boring foundations are what separate a site that runs for years from one that looks great for a month and then quietly breaks. I would rather hand you something smaller that holds up than something impressive that falls over the first time it meets a real customer.
What this looks like in practice
WeWriteRoofs is the clearest example. The founders are roofing estimators who had spent years writing insurance estimates by hand and teaching other roofers to do the same. The expertise was real. The delivery was not, because every report still ran through a person doing it manually, one at a time. There was no way to grow it without hiring more of themselves. I built them a platform where a contractor creates an account, submits a job with photos and a scope form, and gets an organized report back, with the whole history in one place instead of scattered across email threads.
It runs as a web app with native apps for iPhone and iPad, shipped to the App Store in May and updated since. A contractor can capture roof photos on site, let the job address fill in from their location, complete a scope form that covers the insurance, decking, and tarp details an estimator actually needs, then track the status of every project and add their crew to the account. The first contractors are placing real orders through it now.
Most of that market works the other way. It is a done-for-you service. You email a scope form to a stranger, someone writes your estimate, a PDF comes back. Speed is not the hard part there, plenty of those services turn work around in a day or two. The hard part is that the expertise stays locked inside a person, so nothing accumulates and nothing stays organized. Xactimate has a steep enough learning curve that plenty of contractors avoid it altogether and end up underbilling their own jobs. Building this as a product instead of an inbox is the entire point.
Sa1gon Nails and Spa is the small end of the same idea. A nail salon in Schnecksville, Pennsylvania that had not opened yet. Instead of waiting for opening day, we put the site live early with proper local business markup, hours, phone, and a working booking flow, so search engines had months to find them before the doors opened. A local business that launches its website the week it opens is starting from zero on the day it can least afford to.
How I work
I start with the job, not the tech. Before I write anything, I want to know what this is supposed to accomplish and how we will both know it worked. More booked calls this month. Ten hours a week off your admin. Fewer of the same question in your inbox. Then I build toward that, ship a lean first version you can actually use, and grow it from what real people do with it instead of what we guessed in a planning doc.
And if a custom build is the wrong call, I will say so. Sometimes the honest answer is that a simple site, or a tool you already own, does the job. Telling you that costs me a project and keeps your trust, and I would rather have the second one.
If your site should be doing more
If your website should be doing more than sitting there looking finished, that is exactly the kind of problem I like to take on. Tell me what you are trying to get done, and I will tell you honestly whether I am the right person for it and how I would approach it.
So before you brief anyone on colors and logos, it is worth asking the real question. What job do you actually need this site to do?
