Every other 30-day plan says "spend some time each day". This one tells you exactly how long, because that is the thing you need to know before you start.
0 of 30 days · 0 of 15 hours done
Finish four real things and find out whether AI is actually useful for your work. No tools to install, no accounts.
Pick an idea that matches your job and finish it. This is the whole test: does AI produce something you would actually use?
Open this →Run the AI Reality Check. Ask for something it cannot know and watch it answer confidently anyway. Spotting this is the first real skill.
Open this →A different idea from the same cluster. Notice how much less time it takes now.
Open this →Model, prompt, token, hallucination, context. Twenty minutes now saves you from nodding along for months.
Open this →Take your best result and turn the prompt into something reusable. Reusable prompts are the actual product in most paid AI work.
Open this →Pick whichever guide answers the question still nagging you.
Open this →Open My Builds. Four to six real artifacts in a week, from zero. That is your evidence.
Open this →Stop sampling. Choose the single kind of work you will get paid for and build three portfolio pieces in it.
Writing, automation, data work, or building small tools. Pick the one closest to what you already do — existing domain knowledge is what makes you worth more than a beginner.
Open this →Make something in your lane good enough to show a stranger. Save it.
Open this →Work the lesson behind the idea you just used. Now you know why it worked, not just that it did.
Open this →Same lane, different problem. Two pieces beats ten samplers.
Open this →Rework both pieces with what you learned. Keep the prompts — you are building a kit, not one-offs.
Open this →The one you would put first if someone asked what you can do.
Open this →Delete the weak ones. Three strong pieces is a portfolio; nine mixed ones is noise.
Open this →Apply, list, and set up. This is the week most people skip, and skipping it is why month one earns nothing.
Read how paid AI training work actually works before applying to it. Work availability is inconsistent. Treat it as variable income, never a salary.
Open this →DataAnnotation: Assessment must be passed; acceptance is not guaranteed and is the most common failure point. Expect days to weeks — assessment review times vary and are outside your control.
Check the source →One sentence: who you help, with what, and the result. Narrow beats broad — "I write rebooking texts for salons" wins more work than "AI writing services".
Put your one-line offer up as a fixed gig. Fiverr takes a commission, and early orders are usually priced low deliberately to earn ranking. Plan for that.
Check the source →Use your three portfolio pieces. Entry-level $10-$25/hr; intermediate $25-$75/hr; specialised development, AI and consulting $75-$150+/hr — start at the bottom of the range you can live with.
Check the source →Five specific ones beat fifty generic ones. Reference something real in each posting or do not send it.
Reread your profile and gig as a buyer would. Is it obvious in five seconds what you do?
Turn one-off wins into a routine, and be honest about where you actually are.
If something landed, do it properly and over-deliver slightly. If nothing landed yet, that is normal — keep going.
Volume with specificity. Most first contracts come after more attempts than anyone tells you.
Collect the prompts that worked into one document. This is the asset that makes hour three faster than hour one.
Open this →Work a lesson in the track that matches your lane. Depth is what moves you off the bottom rate.
Open this →Ideally something a real buyer asked about. Real requests beat invented samples.
Open this →If anyone has paid you or used your work, ask. Reviews are the ranking currency on every platform.
Which kind of work felt easy? That is your lane for month two.
Open this →An honest audit, not a graduation.
Count what you made, what you applied to, and what came back. Write it down. Vague memories make bad decisions.
Double down on the lane that pulled, drop the one that did not, and set the next 30 days.
Open this →Free forever · no account needed · your progress stays on this device