One of the hardest tasks when using prompts to create AI images is consistency from image to image. For instance if you create a character or person, ideally you want to reuse visual elements of that character but put them in different situations or environments.
About 7 months ago I began to experiment and learn MidJourney, now with the 5th version release, I'm revisiting some of my initial prompts and seeing the new results.
Reflecting on the memorable moments and achievements from the first quarter of 2025.
I really liked the cartoon logo for iDisclose and their LawBot identity. Having the privilege to work with them years ago on a website and an explainer video, it was fun to work on taking their 2D character and bringing it into 3D.
Another interesting project was doing a 3D model from just a series of architectural drawings years before the building actually got started.