Two weeks ago I caught myself saying something I never expected to hear from my own mouth: ‘I just got an extra 30–40 productive hours this week.’
I wasn’t working longer. I was using GitHub Copilot.
What we’re using today— Copilot, Claude, Grok, etc. — is still narrow AI. It’s extremely good at specific tasks (writing code, answering questions, generating SQL). True Artificial General Intelligence (AGI) would be something much more powerful: an AI that can reason across domains, plan long-term projects, and improve itself.
We’re not there yet. But the speed at which even today’s tools are improving makes you wonder how far away that future actually is. GitHub Copilot is upadted once or twice daily.
My daily work covers maintaining and developing SQL applications supported by C# external modules. Before GitHub Copilot, I was navigating between the Microsoft SSMS SQL management tool and Visual Studio. Sometimes the the tools overlapped, sometimes, they conflicted and it was a real challenge to keep both on a limited size screen.
With GitHub Copilot connecting the Github repositories with coding capabilities, I could just ask: "I need to add a control to this UI to select the last dog and poney show in this database". Within seconds, the revised UI was created in C# and the backend query in SQL. Reay for testing. Correct 60 to 70% of the time.
The productivity improvement was an order of magnitude better than the old copy/paste/build/test and ask for fixing issues. GitHub Copilot created the code in my C# repo, checked it and reported success. New SQL queries or Stored Procedures were created and tested in seconds.
It is very hard to fathom that suddenly your working day has 80 hours.
Food for thoughts! And think about how you will beusing the extra time.
