I’ve been a software developer for nearly three decades. In that time, I’ve wired up thousands of API calls, debugged countless integration points, and spent more hours than I care to admit wrestling with boilerplate code that had nothing to do with the actual problem I was hired to solve. That era is over.

Over the past year, AI coding assistants have fundamentally changed the way I work. And I don’t mean they help me autocomplete a line here and there. I mean the entire workflow — the way Fog City Code Works delivers software to clients — has shifted. The result is roughly five times the productivity I had before. That’s not marketing. That’s my real experience, project after project, client after client.

The Old Way

Here’s the dirty secret of software development: a huge portion of the work has always been plumbing. Connecting this service to that database. Translating data from one format to another. Writing the same authentication and error-handling code for the hundredth time. Configuring deployment pipelines. Chasing down a missing semicolon in a config file at 11pm.

It’s necessary work. But it’s not high-value work. None of my clients ever called me and said, “Dave, I really need someone to wire up an HTTP client with proper retry logic.” They called because they had a business problem — a manual process that was killing productivity, a data bottleneck, an opportunity to scale — and they needed software to solve it.

The gap between what I was hired to do and how I actually spent my time was always frustrating. I’d estimate that in the old world, 60-70% of my time was spent on implementation, plumbing and only 30-40% on the part that actually mattered — designing the solution, understanding the business problem, and making strategic technical decisions.

The Shift

AI coding assistants flipped that ratio. Today, I spend the vast majority of my time on architecture, system design, and solving the actual business problem. The repetitive implementation work — the API wiring, the data mapping, the boilerplate — gets handled by AI, with me directing and reviewing.

This isn’t about cutting corners. I’m still deeply involved in every line of code that ships. But instead of typing it all out keystroke by keystroke, I’m operating at a higher level. I describe what needs to happen, the AI generates the implementation, I review it, refine it, and move on to the next decision. The quality is the same or better. The speed is transformational.

The result is that I spend my time where it matters most — on the problems my clients actually hired me to solve.

A New Kind of Superpower

The productivity gain is real, and it compounds. When implementation takes a fraction of the time, I can explore more options. I can prototype three approaches and compare them instead of committing to one and hoping for the best. I can experiment with a new technology in hours rather than days. I can refactor fearlessly because the cost of making changes has dropped dramatically.

This changes the economics of every project. More iterations. More polish. Better solutions. Delivered faster.

Think of It This Way

Imagine an architect who spent years designing beautiful buildings but had to lay every brick himself. One day, he gains a skilled crew that can execute his designs with precision. He doesn’t stop being an architect — he becomes a better architect because he can finally see his vision come to life without being buried in bricklaying.

That’s what happened to software development.

Or think of a master woodworker who spent decades with hand tools. When power tools came along, the craftsmanship didn’t go away. The artistry, the eye for detail, the understanding of wood grain — all of that still mattered. But the ability to execute went through the roof. Projects that took weeks now take days. The craftsman didn’t become less skilled. He became more capable.

It’s like a pilot with autopilot handling the routine portions of a flight. The pilot isn’t less important — the pilot is more important because they can focus on navigation, weather, fuel management, and the big decisions that determine whether everyone arrives safely. The routine work gets handled so the expert can focus on expertise.

What This Means for Clients

This is where it gets exciting. For Fog City clients, the benefits are immediate and tangible.

Faster delivery. Projects that used to take months can now be delivered in weeks. The implementation phase compresses dramatically while the quality of the design work actually improves because I have more time to think about it.

Broader capability. I can now confidently take on projects that span a wider range of technologies. Need a custom desktop application backed by a cloud database with a real-time dashboard? That’s not three separate specialists anymore. AI assistants give me the depth to move across technology boundaries with confidence.

More experimentation. When prototyping is fast, we can try more ideas before committing. Clients get to see working prototypes early and provide feedback before we’ve invested weeks of effort. This leads to better outcomes and fewer surprises.

Better solutions. More time on architecture and design means the end product is more thoughtful, more scalable, and more aligned with the actual business need. Less time fighting with plumbing means more time making sure the solution is right.

The Bigger Toolbox

At Fog City Code Works, I’ve fully embraced agentic AI coding workflows. This isn’t a novelty or an experiment — it’s how I work every day. AI is integrated into my development process from design through delivery. It’s the biggest expansion of capability since the cloud revolution, and I was an early adopter of that, too.

The toolbox is bigger than it’s ever been. Cloud technologies brought enterprise-grade infrastructure to small businesses. Serverless eliminated the overhead of managing servers. And now AI coding assistants have eliminated the overhead of repetitive implementation. Each wave has allowed me to deliver more value to clients at lower cost and higher speed.

I’ve never been more capable of solving complex business problems than I am right now.

Looking Ahead

We’re still in the early innings of this shift. AI coding tools are getting better every month. The workflows are becoming more sophisticated. The range of tasks that AI can handle effectively keeps expanding.

For Fog City Code Works, this means one thing: more innovation and more business value on every project. I’m spending my time where it counts — understanding your business, designing the right solution, and making sure the technology serves your goals. The plumbing takes care of itself.

If you’ve been thinking about a software project but worried about cost or timeline, now is the time to have that conversation. The economics have changed. What was once a six-month project might be a six-week project. What was once out of reach might be well within it.

Contact Fog City Code Works today, and let’s talk about what’s possible.