I'm a marketing and events student, not a developer. I don't have a strong coding background, and I've never claimed to. But over the past year or so I've built a genuine daily habit around AI tools — not because they're trendy, but because they consistently help me work faster and think more clearly on problems that would otherwise take me much longer.
This page isn't a showcase of how AI does everything for me. It's a more honest account: what I actually use these tools for, what I've learned about their limits, and how I've started to figure out when to lean on them and when not to.
My day-to-day rotation is mainly three tools: Claude, ChatGPT, and Gemini — and I've ended up using each one slightly differently based on where they perform best.
Claude is where I do most of my writing and structured thinking. Long documents, case study drafts, content briefs — anything where I need the output to feel coherent and well-reasoned over several paragraphs. ChatGPT I tend to use for quicker creative tasks or when I need a fast answer with a different angle. Gemini has been useful for tasks that touch Google products directly — Sheets, Docs, summarising Drive content.
I don't treat any of them as a single source of truth. I use them as a first draft engine, a thinking partner, or a technical shortcut — and then I edit, rewrite, and make decisions myself.
Content creation & writing
Drafting communication briefs, writing project descriptions, structuring reports and presentations. AI doesn't replace the thinking I still need to know what I want to say but it dramatically shortens the gap between a rough idea and something usable. The editing pass is where the real work happens.
Google Sheets automation — restaurant tips calculator
One of the more concrete examples: I work part-time in a restaurant and the tip distribution at the end of each shift was done manually, which was slow and occasionally produced errors. I used Claude and ChatGPT to help me build a Google Sheets tool that automatically calculates each team member's share based on hours worked, role, and total tips collected. I couldn't have written the formulas from scratch but I could describe exactly what I needed, iterate on the output, and fix the parts that didn't work. The sheet has been in regular use ever since.
This website
This portfolio is probably the most visible output. I had a clear idea of what I wanted the structure, the sections, the overall feel but not the technical skills to build it alone. I used Claude heavily throughout: generating the HTML/CSS base, debugging layout issues, writing the JavaScript interactions, and iterating on components as the design evolved. I directed the process, made the creative decisions, and reviewed every output. The site exists because of that collaboration, but the vision and the content are mine.
Research & synthesis
For school projects audience research, competitive analysis, summarising long documents AI tools cut down the time it takes to get oriented on a new subject. I use them to get a first map of a topic quickly, then go deeper in the areas that actually matter for the brief.
The outputs are only as good as the inputs. If I don't know what I want, the tool can't figure it out for me — it'll produce something plausible-sounding that misses the point entirely. Early on I spent a lot of time fixing outputs that weren't wrong exactly, just not right. That was usually a problem with my prompt, not the model.
AI also has no sense of context unless you give it one. It doesn't know the client, the event, the audience, or the constraint that changes everything. Every time I skip that step and go straight to generating, I get something generic. The discipline of explaining the situation properly before asking for help is something I've had to consciously build.
And for some things it's simply not useful: a strategic decision, a judgment call about tone, a creative direction that needs to feel personal. Those still require a human to make the call.
The biggest shift wasn't learning how to use the tools — it was learning how to think more precisely. To get something useful out of an AI, you have to be able to articulate exactly what you need, what the context is, and what "good" looks like. That forces a kind of clarity that makes you better at the underlying task too, not just better at prompting.
I've also stopped thinking about AI as a shortcut and started thinking about it as a collaborator with a very specific skill set. It's fast, tireless, and great at structure. It has no instincts, no taste, and no understanding of what actually matters in a given situation. Knowing which problems to bring to it and which ones to solve yourself is what makes the difference between saving time and wasting it.