Key facts
- Tech workers are spending evenings and weekends learning AI tools to remain competitive.
- AI tools are seen as increasing productivity, but require significant after-hours learning.
- An Ernst & Young survey indicated 85% of US desk workers are learning AI outside of work.
- Hiring for AI engineers has increased significantly, while other engineering roles have stagnated.
- Some workers are investing personal funds in AI tools, subscriptions, and workshops.
Tech workers are increasingly dedicating evenings and weekends to learning and experimenting with new AI tools, viewing it as essential for career survival in a rapidly evolving industry. Many report spending significant personal time and money on subscriptions, workshops, and personal projects to keep pace with AI advancements.
Maahir Sharma, a software engineer at a Big Tech company, spends about 20 hours a week on AI experiments, noting that hands-on experience is crucial for industry survival. He uses tools like Cursor, which he pays for himself. A survey by Ernst & Young found that 85% of US desk workers are learning AI outside of their regular work hours.
Tanvi Pisal, a product designer, began to worry about job automation due to AI and expanded her skills, only to be laid off months later with the company citing AI adoption. She now dedicates 10 to 15 hours weekly to AI learning and has spent hundreds of dollars on tools and workshops, emphasizing the need to constantly catch up.
While some workers face time constraints, others, like lead engineer Manoj Aggarwal, benefit from employer-provided AI training, allowing them to develop skills on the job. Udit Mehrotra, a head of product at Amazon, approaches AI learning more sustainably, viewing it as a marathon. Amazon itself provides internal AI training resources.
However, for many, the pace of change necessitates intensive after-hours learning. Abhinav Bohra, a senior applied scientist at Amazon, spends eight to 12 hours weekly on AI, investing thousands of dollars annually in tools and conferences. He describes this as a 'learning tax' that blurs professional and personal time, with the primary concern being technical obsolescence in a field with a constantly shifting baseline.
