I get asked this question at least three times a week: "What AI tools do you actually use?"

Not what I've tried. Not what looks cool on Twitter. What I actually open every single day to get product work done faster and better.

So here it is — my full AI tool stack, broken down by use case. Every tool here has earned its spot by saving me real hours, not just promising to.

Research & Discovery

Perplexity — My Default Search Engine

I barely open Google anymore. Perplexity is where I start every research task — market sizing, competitor features, technology trends, you name it. The killer feature? It cites its sources, so I can verify claims before putting them in a deck.

How I use it:

Claude — The Deep Thinker

When I need nuanced analysis, Claude is my go-to. It's particularly strong at synthesizing large amounts of information and finding patterns I might miss. I use it for user interview analysis, strategy brainstorming, and any task that requires careful reasoning.

Writing & Documentation

ChatGPT — The Versatile Workhorse

For drafting PRDs, user stories, and stakeholder communications, ChatGPT with GPT-4 is hard to beat. I've built custom GPTs for specific tasks:

Pro tip: The custom GPT feature is underrated. Build one for your most repetitive writing task — you'll save 20 minutes every time you use it.

Notion AI — Inline Documentation Help

Since our team already lives in Notion, having AI built right into the docs is a game-changer. I use it for summarizing meeting notes, cleaning up rough drafts, and generating action items from long discussion threads.

Design & Prototyping

Figma AI — Design Feedback at Scale

I'm not a designer, but I review a lot of designs. Figma's AI features help me annotate prototypes faster and generate layout suggestions. When I need to mock up a quick wireframe for a meeting, AI-assisted design gets me 80% there in minutes.

v0 by Vercel — Instant UI Prototypes

This one's newer in my stack but I'm already hooked. Describe a UI component and v0 generates working code. I use it to create quick prototypes during discovery — nothing fancy, just enough to show stakeholders "here's roughly what I'm thinking."

Data & Analytics

ChatGPT Code Interpreter — Instant Data Analysis

Upload a CSV, ask a question, get a chart. I use this constantly for:

It's not going to replace your data team, but it'll make you 10x faster at exploring data on your own.

Automation & Workflow

Make.com — Connecting Everything

The glue between all these tools. I have automations that pull competitor updates into a Notion database, summarize Slack threads, and generate weekly digest emails. More on this in a future post.

What I've Tried and Dropped

Transparency matters, so here's what didn't stick:

The Bottom Line

My stack isn't about having the most tools — it's about having the right tool for each job. Start with one or two that address your biggest time sinks. Master those before adding more.

The PMs who'll thrive aren't the ones using the most AI tools. They're the ones who've figured out which tools genuinely accelerate their specific workflow.

What's in your stack? Hit reply and let me know — I'm always testing new tools.

Need help building your AI-powered workflow? At Spicy Advisory, we help startups integrate AI into their product and GTM processes. For enterprise teams, check out our AI adoption programs.