DevHawk Origin Story

DevHawk Origin Story: Oh, the days of 8 week sprint cycles at AT&T... perhaps 3 weeks of which was work!

I was at AT&T as a contractor for a couple of years, helping automate some of their broadband internet diagnostics systems.

This was AT&T Management's view of a sprint:

• 1 week of planning / architecture
• 4-5 weeks of development, with 2-3x / week check-ins by project managers
• 1 week of integration work to bring all the code together
• 1-2 week of testing/fix feedback loop, and then deploy

And this was Developers' view of a sprint (the reality on my team):

• 1 week of trying not to fall asleep through a week of planning
• 4 weeks of trading stocks, playing fantasy sports, or showing up an hour a day (all real examples!)
• Just tell the PMs we're making progress, offer vague updates
• 2 weeks of actual writing code
• 1 week of testing/fix feedback loop, and then deploy

It's probably not this bad most places, but I know this resonates - because I still see signs of this behavior today!

Time tracking fixes exactly 0% of this.

You have to track real productivity, and PMs need help getting devs past real blockers, or if they're pulling an AT&T, just spurring them to work!

Can an agentic project manager help? We're starting to see the results...!