Practical AI that solves real problems
There's no shortage of AI hype right now. Everyone's got a chatbot, every SaaS has "AI-powered" in the tagline, and half the pitches out there are solutions looking for problems. I take a different approach. I build with AI every day, in production code, client projects, and my own business operations. I can help you figure out where it genuinely adds value and where it's just noise.
What I build
AI is a broad term that covers everything from language models to computer vision. I focus on the areas that deliver the most practical value for small and medium businesses. The kind of work where AI genuinely saves time, reduces errors, or creates capabilities you couldn't have before.
Custom AI assistants & chatbots
AI assistants trained on your specific content, including your documentation, product information, FAQs, and knowledge base. Embedded directly on your website, they answer customer questions accurately using your actual information, not generic responses. They handle the repetitive enquiries so your team can focus on the conversations that matter.
AI workflow automation
Connecting AI to your existing tools to automate the manual work that eats up your day. Form submissions that get intelligently categorised and routed. Emails that are triaged and drafted. Documents that are processed and filed. I build automations that slot into your current workflow, not ones that require you to change how you operate.
Document & data processing
AI-powered extraction, classification, and routing for your documents and data. Invoices, applications, reports, and forms. If your team is spending hours manually reading, categorising, and entering data from documents, AI can handle the heavy lifting. I build systems that extract the information you need and put it where it belongs, with human review where it matters.
AI-enhanced web applications
Adding intelligent features to your existing web applications and websites. Smart search that understands intent, not just keywords. Content recommendations based on user behaviour. Automated tagging and categorisation. These are targeted AI enhancements that make your existing systems significantly more useful without requiring a complete rebuild.
AI strategy consultation
Not sure where AI fits in your business? I can help you figure that out. I'll look at your operations, your pain points, and your goals, and give you an honest assessment of where AI would genuinely help and where you'd be better off with traditional software solutions. No sales pitch, no inflated promises. Just a practical roadmap based on what I've seen work in real businesses. Sometimes the answer is "you don't need AI for this", and that's a valuable answer.
Most AI projects fail because they start with the technology instead of the problem.
I start with what's actually costing you time, then figure out whether AI is the right solution.
How I approach
AI integration
I don't bolt AI onto things for the sake of it. Every integration I build follows the same principle: start with the problem, build something reliable, and only bring in AI where it genuinely outperforms a simpler solution. Here's how that plays out in practice.
Start with the problem
Before touching any AI tools, I need to understand what's actually costing you time, money, or quality. What are the bottlenecks? What manual processes are draining your team? What would make the biggest difference if it just worked better? The clearer the problem, the better the solution.
Identify where AI actually fits
Not every problem needs AI. Some things are better solved with a well-designed form, a database query, or a simple automation. I'll be honest about where AI adds genuine value versus where traditional software is faster, cheaper, and more reliable. You only pay for AI where it earns its keep.
Build reliable systems first
AI is probabilistic. It doesn't always give the same answer twice. So I build deterministic, reliable software as the foundation, then layer AI on top where its flexibility is an advantage. Error handling, fallbacks, and monitoring are built in from the start. The result is something you can actually depend on.
Ship production-ready work
I don't build demos. Everything I deliver is production-ready, properly tested, documented, and designed to run reliably in your business. I'll make sure your team knows how to use it, set up monitoring so you know it's working, and provide ongoing support to refine things as you learn what works best.
Who this is for
AI integration isn't just for tech companies with big budgets. Some of the most effective AI applications I've seen are in small businesses where a targeted solution removes a significant pain point. Here's who I tend to work with on these projects.
Business owners curious about AI
You've seen the headlines and you know AI is changing things, but you're not sure where to start or who to trust. You want someone who'll give you a straight answer about what's realistic for your business, not a sales pitch wrapped in jargon.
Teams drowning in manual data processing
Your team spends hours each week reading documents, copying data between systems, categorising enquiries, or doing other repetitive knowledge work. You know there has to be a better way, and there probably is.
Businesses wanting smarter customer interactions
You want to offer your customers instant, accurate answers from your own knowledge base without hiring more support staff. A custom AI assistant trained on your content can handle the routine questions while your team focuses on the complex ones.
Anyone who's been burned by AI hype
Maybe you've already tried an AI tool that promised the world and delivered a glorified autocomplete. Maybe you've sat through pitches full of buzzwords and empty demos. I get it. My approach is the opposite: start with your actual problem, be honest about what AI can and can't do, and only build what's going to work.
Why work
with me
I'm not an AI consultant who read a whitepaper and built a landing page. I use AI in production every single day.
My development practice is AI-augmented from top to bottom. I use AI for code generation, architecture planning, testing, documentation, data processing, and client communication. I've built AI-powered features into client systems. I run AI agents that manage parts of my own business operations. This isn't theoretical knowledge. It's daily, hands-on experience with what works and what doesn't.
But here's the thing that matters more: I've been building software for nearly 20 years. I know how to build systems that are reliable, maintainable, and actually solve the problem they're designed to solve. AI is an incredibly powerful tool, but it's still a tool. It needs to be integrated into well-architected software by someone who understands both the capabilities and the limitations.
That combination of deep, daily AI practice plus 20 years of engineering judgement is what I bring to every project. You get someone who can build the AI integration and the production software it runs inside. No handoffs, no gaps, no one blaming the other team when something doesn't work.
Frequently Asked Questions
A custom AI chatbot trained on your content typically starts from around $3,000–$5,000 for the initial build, depending on the complexity of your content, the number of data sources, and how deeply it needs to integrate with your existing systems. There are also ongoing costs for AI API usage, which vary based on how much the chatbot is used, but for most small businesses, this runs to tens of dollars per month, not hundreds. I'll give you a clear cost breakdown before we start so there are no surprises.
Not if it's implemented well. The most effective AI integrations I build are designed to handle the repetitive, time-consuming tasks that prevent your team from doing their best work, things like sorting enquiries, extracting data from documents, or answering routine questions. The goal is to free your people up for the work that actually requires human judgement, creativity, and relationships. AI is a tool that makes your team more effective, not a replacement for them.
I work across the major AI platforms including OpenAI (GPT-4), Anthropic (Claude), and open-source models where they make sense. The choice depends on the specific task: some models are better at conversation, others at data extraction, others at reasoning through complex problems. I also build with vector databases for knowledge retrieval, workflow automation frameworks, and custom API integrations. I'm not locked into any single vendor. I use whatever works best for your situation and budget.
