Activity
Mon
Wed
Fri
Sun
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
What is this?
Less
More
420 contributions to AI Automation Society
Day 1 - Newsletter (GTA VI)
Built my first AI newsletter automation today. The first graphics were awful. Generic, flat, obviously AI. So I stopped describing what I wanted and showed it instead. I had it study the official GTA 6 website for its colors, fonts and style. The second try looked incredible. Same tool, same cost. Lesson: AI is only as good as what you give it. Don't describe it. Show it. Next Time: I would like to add more in game graphics! But I am happy with my first run.
Day 1 - Newsletter (GTA VI)
1 like • 6h
Nice graphics. He pushes further. It's a good start 🔥🦾
#7dayAISChallenge - Day 3 ✅
This was the most inspiring day yet, i realised that i can build my own JARVIS. I was thrilled with how this personal assistant can 10x my productivity. So, i built a school-presentation skill, it creates simple educational presentations on certain topic just from page photos. Everything you need to do is to just attach your page photos of the topic you want and hit /school-presentation and it creates you cool presentations. You trigger it by slash command or just natural command. After initial run, i did 3 feedback loops - and achieved great progress. The format and style of presentation fit what i have wanted - intro in the first slide, and questions about the topic at the end - and it become consistency. Now, i am up for building my own JARVIS
#7dayAISChallenge - Day 3 ✅
1 like • 6h
Great job. Leading in the next days of the challenge 🔥🙏🦾
#AISChallenge - Day 2 successfully completed
This one took a bit longer than I expected. The integration with Codex went smoothly and it performed very well; the main issue was that I lost track of token usage and ended up consuming way too much, haha! 1 - What I extracted: I ran an initial round of tests just to understand how to make requests and get familiar with the syntax. Afterward, I extracted contact information for aesthetic clinics in my city via Google Maps (I didn’t open it directly; I simply made the request, and Codex provided the list). 2 - What I learned: I really liked that it notifies you how many credits will be required to execute a search. I found the credit consumption to be quite low, considering the initial results. Not bad at all. 3 - A use case: Definitely lead research and enrichment for prospecting. I was impressed by the sheer amount of information I could request. Overall, I’d say it was very productive. Now, I’m testing n8n’s MCP. That one is proving to be a bit tricky, as I’m running it locally via Docker. Plus, the token consumption is absurd - I’ve already hit the limit twice! Just waiting for it to reset so I can continue, haha!
#AISChallenge - Day 2 successfully completed
0 likes • 6h
Lats goo🔥🔥🔥
Day 3 done ✅ #AISChallenge
What I built: an infographic mode inside my own video skill (reeler). I trigger it with "make an infographic from report X" or /reeler. It reads a markdown report, writes the numbers and short texts into a small JSON, and renders a PNG from HTML in local Chrome. Same idea as Nate's infographic builder, with one change from the start. I need the numbers exact, so the image comes from code instead of an image generator. The AI-generated version is still there on request, but every digit and letter gets checked before it's used. One optimization after watching it work: a guard. Before rendering, a script checks that every number in the JSON exists in the source report. On the first real run it stopped the two totals, because they don't appear in the report as such. Now a total only passes if the sum of its column actually adds up. That check runs as a script, so it doesn't depend on the model getting it right. Six runs so far, and watching them showed me what to fix: an orphan card in the calendar, a sliver on zero-value bars, long names wrapping, and the bottom line cut off because the size was measured before the font loaded. For Dive Deeper I added the MCP part: the same JSON goes to ChatGPT through Claude in Chrome, and the generated image comes back to be checked against it. First try, GPT also chose to build it from code. When I asked for the image generator, every number and word came out right, but the bars were only roughly proportional and the size was 1024×1536 instead of what I asked for. Both images below, same data: code on the left, GPT on the right.
Day 3 done ✅ #AISChallenge
0 likes • 6h
Congratulations on doing a good job 🙏🙌
AIS 7 Day challenge Day 1
I did it! OK Claude did it with my input. What I built: I built a newsletter that matches my brand and does this automatically Monday morning. One thing you learned: What the env file is for. One thing you'd improve: better content
AIS 7 Day challenge Day 1
1 like • 6h
Amazing start🙏
1-10 of 420
Piotr Lewandowski
7
5,120 points to level up
@piotr-lewandowski-7838
AI agent Maker

Active 6h ago
Joined Jul 3, 2026
ISTJ
Powered by