I get meeting scheduling done with Google Calendar and a shared link instead of a booking page, because in a team of a handful of people the calendar already shows who is free and a link avoids a round of emails. I put every call on a 25 minute slot with 5 minutes after it, and I keep a separate calendar for focus blocks so nobody can book over the hours when I actually ship code. The part that took the longest to fix was time zones: I always write the zone in the invite text now, because one silent mismatch cost us a call with nobody on it.
I get AI writing done by using ChatGPT to draft the part I already understand, never the part I do not. For our tool's help text and FAQ I paste in the real features and limits first, ask for three short versions, then rewrite the one I keep in my own words and cut every claim I cannot stand behind in the product. It is good at turning a list of facts into sentences and bad at knowing what the tool actually does, so the facts have to come from me. The step that matters is the last one: reading it out loud and deleting anything that sounds like an advertisement.
I get note taking done with Apple Notes and a plain GitHub repository for anything that has to survive. Apple Notes wins for the fast parts, because I can drop in a link or a screenshot from the share sheet mid-task and search it later with one word I half remember. The part that matters most in a small team is where notes stop being private: anything a second person needs to read goes into a Markdown file in the repository next to the code, so the decision sits beside the change it explains. Keeping both means the scratchpad never becomes the only copy of anything important.
I get photo editing done with Snapseed on my phone for anything quick, because it does the three corrections I actually need (straighten, exposure, white balance) without an account or an upload step, and the selective brush lets me darken just the sky. For the files that end up on a product page I take them into GIMP on the laptop instead, so I can set an exact pixel size and check the result at 100 percent. What saves me the most time is doing the quick pass on the phone first and only opening the desktop editor for the shots that are going to be published.
I get image editing done with GIMP for anything that needs a proper job: odd background removal, cleaning a scan, or resizing a batch of product shots. The part that earns its keep is the export dialog, where I can set the exact pixel size and keep the alpha channel, so the file is ready for the site without another pass. It is slower to learn than the web editors I keep in a tab, but it never asks me to sign in or upload anything, which is why it stays open on my machine.
I get it done with the browser's own DevTools Console and the deploy logs in our hosting dashboard, because together they tell me whether a problem is inside the page or in the build that shipped it. When a report says a button did nothing, the Console usually shows the exact error and file; when the page loads but a feature is missing, the deploy log shows a failed build step. Checking both first saves me from guessing.