Sign up to save tools and stay up to date with the latest in AI
bg
bg
1

Ask HN: Daily practices for building AI/ML skills?

Dec 14, 2023 - news.ycombinator.com
The article suggests a daily one-hour routine for developing AI/ML skills. It recommends spending 30% of the time (20 minutes) reading textbooks, blogs, and newsletters to build foundational knowledge, with sources such as Hands-On Machine Learning with Scikit Learn & TensorFlow, Towards Data Science, Analytics Vidhya, and Monica Rogati's newsletter. Another 30% of the time should be spent on practical online courses on platforms like Coursera, Fast.ai, and Udacity, focusing on those that provide labs and exercises.

The remaining time should be divided between working on small projects that interest the learner (30%) and engaging in AI/ML communities like Reddit's machinelearning and Kaggle forums (10%). The article emphasizes the importance of combining theoretical knowledge with practical implementation and encourages learners to gradually take on more ambitious projects and compete at Kaggle. It concludes by encouraging patience, persistence, and a willingness to continually iterate on one's skills.

Key takeaways:

  • Spend 30% of your daily AI/ML development time reading textbooks, blogs, and newsletters to build foundational knowledge.
  • Allocate another 30% of your time to doing practical, hands-on online courses.
  • Use 30% of your time working on small projects that interest you, translating what you're learning into code.
  • Engage in AI/ML communities for 10% of your time to stay up to date with the latest developments.
View Full Article

Comments (0)

Be the first to comment!