Josh Longenecker
@jlongen.bsky.social
AI/ML @ AWS
Opinions are my own
Opinions are my own
Pinned
Josh Longenecker
@jlongen.bsky.social
· Sep 6
I mentally mapped out the genAI space into 3 “layers”: Modeling, Serving, and Consumption.
I’m quite curious to know which layer most focus on, or if you’re insane like me and TRY to be great at all 3 🥲
I’m quite curious to know which layer most focus on, or if you’re insane like me and TRY to be great at all 3 🥲
I mentally mapped out the genAI space into 3 “layers”: Modeling, Serving, and Consumption.
I’m quite curious to know which layer most focus on, or if you’re insane like me and TRY to be great at all 3 🥲
I’m quite curious to know which layer most focus on, or if you’re insane like me and TRY to be great at all 3 🥲
September 6, 2025 at 10:19 PM
I mentally mapped out the genAI space into 3 “layers”: Modeling, Serving, and Consumption.
I’m quite curious to know which layer most focus on, or if you’re insane like me and TRY to be great at all 3 🥲
I’m quite curious to know which layer most focus on, or if you’re insane like me and TRY to be great at all 3 🥲
I decreased my flux kontext diffusion pipeline latency by 30%, so I wrote a blog to show you how.
Full breakdown: medium.com/@jldevtech/m...
Full breakdown: medium.com/@jldevtech/m...
Making Flux Kontext go brrr
Speed Up Your Diffusion Models: 30% faster inference with torch.compile
medium.com
September 6, 2025 at 6:15 PM
I decreased my flux kontext diffusion pipeline latency by 30%, so I wrote a blog to show you how.
Full breakdown: medium.com/@jldevtech/m...
Full breakdown: medium.com/@jldevtech/m...