sorry for the delay in getting this post summary out! I made it last week, but was behind on posting the summary here. My process is fairly straight forward and it comes down to two key pieces:
1) get a good starting frame (and ending frame if you want to predict outcome) - I use GPT Image2.5 as its the leading ai image model
2) then prompting correctly, which seems easier than it actually is. These models all act a specific way. So learning how to use camera angles and imagery descriptions really does help
The guide has the full walkthrough of my process, so feel free to play around! would love your opinion on it too! Kevin