“Can ML solve my problem?

What it takes to get to the level, ML Perspectives

Even answering the question “Can ML solve my problem?” requires you to overcome half of the challenges ML libraries, Find state-of-the-art (SOTA) deep neural networks, experiments, and MLOps.

Machine learning often boils down to the art of developing an intuition for where something went wrong.

Certain behavior signals with where the problem likely is in your debugging space - preprocessing, data issues, optimization, weak labels, and learning rates. ML Learning = Experimentation = Experience.

The efforts in terms of marking production-ready apps - Text, Code copilot are in (primetime)

Vision tools/custom tools are evolving rapidly. The maturity of imagegen, DALLE3, and Midjourney is nearing prime time in 2024, One area is learning to build production solutions where there is maturity. Another area is building custom tools where maturity is nearing prime time :)

Ref - Link1, Link2

Keep Exploring!!!