Hello everyone, Sumit here from the Himalayas. I hope everyone is having a relaxed weekend. My workflow with vibe coding has settled pretty well as I get more and more time out of desk while Claude Code builds the software. I wanted to offer any assistance to fellow founders. I have been vibe coding full-time for a little over 4 weeks. Wrote about it here. Please share your tools, or workflow and in particular what is not working for you. What is frustrating you in building software with vibe coding?
I used to plan my lunch breaks around AI tasks. Pick the one task that's easy enough to just hand off, start it, go eat.
Came back more than once to find the agent had stopped two minutes in over some dumb question. Not even a permissions thing, it just sat there waiting for me to say "go ahead". Didn't even know it was stuck until I checked.
I have a question for all of you specially to those that vibe code and use different Ai tools and are non-technical. How do you know your product is secure and safe? As most Ai tools have their guardrails triggered when they hear the word security and how do you know or find out the code is complete, for instance if you can t find a human reviewer do you trust the Ai then?
Solo builder here, using Cursor + Claude Code as basically my whole engineering team. Six months in I still don't have a good instinct for this: some days I let it just run with a whole feature and barely glance at the diff, other days I'm rereading every line like it's a junior dev's first PR.
Right now my rough rule is billing/auth code gets the paranoid review, UI and analytics panels get the "looks fine, ship it" treatment. But that's just a gut feeling, not a real system.
Anyone here have an actual framework for this, or is everyone just winging it based on vibes and how burned they've been before?
I build solo, with coding agents doing most of the typing. Going through this week's threads here, I noticed the bugs that cost me the most had one thing in common: nothing crashed, nothing logged, and no user would ever have reported them.
1. Tests that couldn't see the risk. My API's pytest suite runs on SQLite, which has no Postgres row-level security or triggers. Green every time. Now I check RLS by hand with psql, as anon and as a logged-in user, inside BEGIN...ROLLBACK.
For a while I ran everything on the strongest model: hard design work, but also quick emails, translations and where is this function defined? searches. It worked, but most of those tasks didn t need it, and every side task made my main session longer and slower.
So I set up three subagents, each with its own model:
An email and Slack drafter on Haiku
A language translator on Sonnet, since translation needs more nuance
A read-only codebase explorer on Haiku that returns a short report instead of dumping files into my main session
Design, debugging and final review stay on the strongest model.
We run an automated pipeline that drafts articles for our product. During our first test attempts we let the model do almost everything, including the mechanical bits: fixing dashes and quotes, tracking which stage each article was in, creating SEO json, naming image files etc. It worked, but the quality was uneven, sometimes the model would miss some things, and, of course, it was rather expensive in terms of tokens. Then we changed our approach and started treating AI like a creative employee and providing it with "software" (python scripts) to help with any mechanical tasks that could be automated by code. The model only does the part that actually needs intelligence and creativity now. And honestly, the result surprised me - quality went up, because we now let it focus on what it's actually good at. As a pleasant bonus - the costs of running the pipeline dropped too. One of the first things we scripted was stripping the em-dashes the model loves to add but my co-founder has an allergy for
How do you improve the quality of your AI results, especially automated pipelines?
