3 ms·
My experience has been quite different. At Amazon, after year one, I didn't learn much that was new - just a lot of the same stuff month-over-month. I found th
by seadan83 3y ago
My experience has been quite different.
At Amazon, after year one, I didn't learn much that was new - just a lot of the same stuff month-over-month. I found the biggest growth was when I switched jobs and learned a new environment, systems, team processes; and that was second to working on open source where I probably learned at about 5x the rate I was compared to professional work.
On each job hop, every 2 to 3 years, the pay increase was 30% to 50% each time. I'm now at year 5 in my current job, I've gotten a 10% pay raise just based on the company giving raises, and got a 20% raise in a counter-offer.
Thus, job hopping, I learned a lot more & was effectively increasing my pay by 10% every year. Now that I'm at the same place for 5 years, my pay effectively increased by less than 5% per year and that number is dropping (I'm at max salary, no raises are forecasted for me).
I think it's largely situational what is better for someone. I've known various "senior" MS or Amzn engineers that had long since been so far removed from coding that they were lousy developers (10 years of the same two years of experience). Thus, I think growth (salary and personal) is largely just what you do with it, IMHO there are likely bigger variables at play compared to job tenure.
- elevatedastalt 3y agoI agree that it's definitely nuanced. However, I have also seen many cases of people "failing upwards". They basically do a mediocre job at their current company with middling performance ratings, such that there is no way they would get a promotion (for obv reasons), then jump to another company at L+1, rinse and repeat, until they plateau after 2-3 promotions. It's good for them personally, but terrible for the industry, as these people with actually very mediocre experience and ability are now leads.