non-technical founder here. i spent a year asking my outsourced dev team for repo access. when i finally got in, deploys had been silently failing for 8 days
retained text
started my b2b saas in 2020. i'm not technical, so an outsourced team built it and i stayed out of the code. real customers, seed money, the usual.
the thing that broke it for me was tiny. i asked for a monthly email to customers showing what they'd built with us that month. data was already in the db. loom does it, duolingo does it. i asked for months and it never shipped. when a team can't do the easy thing, you learn something about the team.
so i pushed my way into our own github this april and started reading it with claude doing the explaining. first weekend, it pointed at a column on our production containers called AGE that said 8 days. turns out every deploy for over a week had been failing because the config still pointed at azure after we'd moved to google cloud. one old container was holding up the whole company. two days earlier i'd seen a symptom of it and confidently blamed something else.
same week we found a couple of places where one customer could change another customer's settings, one of them live since jan 2023.
first 26 days on my own: 30 changes to production, 25 bugs closed, and 16,340 lines of dead code deleted in one go. the monthly email took an afternoon.
it's not a rocket ship and there's still a long bug list and no automated tests, which scares me more than anything above. but i sleep better than i have in years.
if you're a non-technical founder with an agency: do you actually have admin on your own repo and cloud account? i didn't, and i should have from day one.document fetched 2026-10-31 00:28 · canonical https://reddit.com/r/SaaS/comments/1xjfvgs/nontechnical_founder_here_i_spent_a_year_asking