People sometimes ask whether AI makes me nervous as an engineer. Honestly, it is the opposite: it is the most exciting thing that has happened to software in my 11+ years in tech. Not the hype, not the demos that go viral on a Friday - the real thing underneath: software can finally work with language, documents, and voice the way people do.
What I love first is the new kinds of products. A few years ago, “read this messy PDF and fill in the form” or “answer a customer from our own knowledge base, with a source” was a research project. Today it is a sprint. A text-to-speech pipeline can turn a book into an audiobook; a RAG system can make a company’s documents answer questions; an agent can take a boring multi-step process and run it, with a human approving the last step. Whole categories of software that used to be impossible are now just engineering.
Then there is leverage. I work hands-on, without an agency or a bench of juniors behind me, and AI tools changed what one senior engineer can deliver. Boilerplate, first drafts of tests, migrations, a quick script to check an idea - the assistant does the typing, I do the thinking and the review. That frees time for the parts clients actually pay for: architecture, product decisions, and the edge cases that decide whether a feature can be trusted.
I also love the engineering puzzle. A model is not a normal function: the same input can give a different output, it can be brilliant and wrong in the same sentence, and every call has a price. Building something reliable on top of that is a genuinely new craft - schemas for structured output, retrieval that finds the right paragraph, evaluations instead of vibes, guardrails that fail closed, logs that explain what happened. Every one of those problems is interesting, and most of them are not fully solved yet.
And I love the ecosystem and its pace. OpenAI and Anthropic APIs, open models you can run on your own server, LangChain and LlamaIndex, the Vercel AI SDK, MCP connecting agents to real tools, vector search inside plain Postgres. AI pushed me into Python, into audio, into ideas I would never have touched otherwise. Every few months the landscape shifts, and I get to be a beginner again - the best feeling this profession has to offer.
What I do not love is AI as decoration: a chatbot on a landing page so the pitch deck has a buzzword, or a model allowed to write to the database with nobody checking. Loving a technology means taking it seriously - cost, privacy, and a human in the loop where the stakes are real. If you have an idea that felt impossible a few years ago, send it to me. There is a good chance it is now just a well-scoped project.
Give the next product someone who stays.
Ready to get started? Send me a message - I'm available to start now and to stay. Long-term retainers, not a two-week handoff. Let's discuss your project, requirements, and the best way to move forward.