Improving cold outreach messages with AI

I was already using ChatGPT. I had Google (Gemini) on my phone. My workflow was "working." But after 7+ years in design, I've learned: the tools that change how you think are worth paying attention to. Not the loudest ones. The ones that make you sharper. So I gave Claude a real chance — and wrote about what actually surprised me as a working designer.

In the article I cover:
→ How I use it to articulate design decisions to clients
→ Writing case studies that tell a story, not a feature list
→ Stress-testing presentations before I walk into the room
→ Cold outreach that actually gets replies

If you're a designer still on the fence about AI tools, this one's for you.

#UXDesign #AITools #DesignThinking #ClaudeAI #PratikBarotDesigns #DesignerLife #ProductDesign #PersonalBranding #DesignCommunity #AI

I've been experimenting with AI tools for LinkedIn outreach lately and it's a bit of a mixed bag. The automation stuff definitely saves time, especially for scheduling posts and initial connection messages, but I'm finding the line between efficient and spammy is pretty thin. Seen heaps of people get flagged for over-automating their DMs, so I'm trying to use AI more for ideation and timing rather than full message generation. The native LinkedIn AI features for suggesting post formats and optimal times seem solid, but anything that sounds too generic just doesn't land with people.

Curious what's working for others though. Are you leaning more into full automation for outreach, or keeping that human touch in the actual messaging? And does anyone actually get decent conversion rates from fully automated sequences, or is that just a myth at this point?