honest reflections
Things I'd do differently.
The experience page lists what worked. This page lists what I'd approach differently — not as confession for its own sake, but because the lessons that changed how I work came from these, more than from the wins.
I would have made the organisational decisions more decisively.
At Sheypoor, some technical problems were visible before the organisational structure around them was clear. I under-estimated how much that ambiguity slowed the technical work itself. In retrospect I would have made structural decisions earlier, clarified roles earlier, made the interaction model with stakeholders clearer, and established clearer organisational boundaries sooner.
I would price sunk cost in the business's currency, not the team's.
The custom SSR framework at Sheypoor represented more than a year of somebody's best work — which makes it emotionally hard to retire. The right question was never "how much have we invested", but "what is continued ownership draining from the business right now". I'd ask that question sooner, and more bluntly, next time.
I would treat release cadence as a leadership instrument from day one.
A two-year gap between production releases didn't start the week I arrived — it had been normalised slowly enough that nobody screamed. Cadence is like cash flow: slow is not itself the disease, but it is almost always a symptom that deserves leadership attention long before it becomes a crisis.
This page will grow as I write more. The point is not self-criticism as a genre — it's that credibility about engineering leadership is earned by showing the decisions you'd take back, next to the ones you'd take again.