... heard people talk about using evaluation agents or automated checks but setting those up properly isn't always clear when you just want to ship fast. What's your simple setup or routine for testing and catching bugs when coding with AI tools? Comment from George Basov(@gbasovv): My routine is less about tests and more about making the app tell on itself. Every debug print in the app also goes to a log file on the device, in every ...

... preview environment, redeploy, and re-test by reading the table, not the screen. The tradeoff I keep going back and forth on: swallowing the error is right for the user, and wrong for me. How do you catch failures like this in serverless apps, where the user sees success? Alerts on specific log lines, a health check that writes and reads a test row, tests that query the database after the action, or something else ...

... success payload. The UI told users their setup was complete while the database sat empty. It gets worse on the security side. If the code inside the try is a permission check or token validation and the catch returns 200, every failed check reads as a pass. The agent is optimizing for a valid response shape, so whatever blocks it, like auth middleware, quietly gets bypassed. One cheap guardrail I ve seen: write the negative-case test (the 403) before ... ... happy-path test, so the suite can t be satisfied by deleting the check. I d love to hear how you handle this: Have you hit this? How did you first notice? Do you audit every catch

... promises to update one row, I d have it return the affected-row count or updated row ID from that same write, and treat zero rows as a failure. Then alert when accepted updates and committed updates diverge. That catches this silent no-op without reading every row back after every request ...

... minute check for anyone whose code catches a broad exception around a service call and shows its message to the user. In PHP, every database error I can get is a RuntimeException. So is almost every refusal I throw on purpose: please add at least one item, this plan does not include that. From the first commit my controllers caught RuntimeException around service calls and showed the message, which is right for the refusals and quietly wrong for everything else ... ... payment had permanently failed, from the branch meant for refusals, while the branch that says try again later never ran. On every document page at once, not found, for documents that exist. In the API the same kind of catch

... launches, or is that section mostly reserved for makers and their network? Should I upvote anything that looks cool, or save upvotes strictly for products I would genuinely use myself? Right now I'm just scrolling and clicking whatever catches my eye, but I keep worrying I'm missing some unwritten rule or etiquette. Also, how is everyone catching

... agents. They're good at the task in front of them and they have no opinion about what the repo already does, so every session starts from a blank slate even when the codebase doesn't. Bad code you catch in review. A second correct implementation of something you already have reads fine in a diff, and it only shows up later when a change has to be made in four places. What's helped is a 40 line conventions ... ... HTTP, dates, queues and logging, pasted in at the start of anything structural. New duplicates mostly stopped. It does nothing about the ones already in there. Has anyone found a way to detect these after the fact? Grepping imports catches

... transaction. Comment from Rabnoor Singh(@rabnoor_s): the thing id gate isnt the refund, its the ratio. an agent that can issue refunds is either switched on or switched off, and both settings are useless in practice. what actually catches your example is checking the amount against the order it came from, because 1000 against a 100 order is impossible rather than just large, and impossible is something you can test without a model in the loop. the other half ...

... first question friends asked when I showed them https://convert.softorino.com , and it's fair, free converters usually pay for themselves with ads, tracking, or your files. The honest answer: the catch is that some of you will like it enough to look at the paid Mac apps I work on. That's the whole model, a good free tool as the introduction. Launch is this Sunday (Aug 30), so before then I want to hear it straight: what makes ...

... empty is genuinely correct. Logging every network response, which doubles the log volume and still needs me to know what I am looking for. A manual pass on a real device before every submission, which caught this one and catches nothing that needs a specific account state. So the question I keep coming back to: how do you catch

... Went through a few companies' software subscriptions recently and found the same pattern every time: multiple tools quietly doing the same job, nobody having noticed. How does your team currently catch that kind of overlap a regular review, a specific tool, or mostly just luck? Comment from Hossein Yazdi(@hosseinyazdi): A regular audit helps, but I think the bigger win is keeping a simple inventory of what each tool is actually used for. Overlap becomes much easier to spot when ...

... users eventually upgrade to pro when they outgrow the limits. $19/month from those users covers the free users who stay free forever. standard SaaS math. That's it. That's the whole model. "ok but what's ACTUALLY the catch? Are you selling my data?" No. and I'm going to be specific about this because " we care about privacy " is what every company says right before they get caught not caring about privacy. Your API keys are encrypted ... ... your API usage). No gating basic features behind paywalls. free gets unlimited chat, smart context management, verified skills, Telegram, Slack. Same features as paid, just with usage limits. "What's actually different on paid then?" More tasks (unlimited vs

... correct your agent's output because you don't trust the first draft. Do the same for your own state. A five-minute end-of-day check, what actually happened, what I made up, what I'll do tomorrow, catches the spiral before it compounds, the same way a code review catches

... Before opening a PR, I started pasting my code into Claude for a quick review first, just to catch anything obvious before it goes to the lead. A few times it caught things clearly worth fixing: a missed null case, a variable name that didn't match the rest of the file's convention, a catch

... allowed to do. Then the handler credited the customer twice. Why it slipped through The bug wasn't on any line. It was the assumption between the lines: every event arrives exactly once. Reading the diff can't catch that. My tests didn't either. I wrote the fixtures, so I never thought to send the same event twice. My assumptions were just checking my assumptions. What actually caught it Running the failure. Deliver the event, deliver it again, watch ... ... usually don't. for the ones that slip through, detection's tougher, and for the same reason: your logs and db agree with themselves. a double-credit just looks like two legit credits. only an independent truth source catches

... trip? Airline, hotel, car rental, cruise all stories welcome. I'm genuinely collecting these to keep improving what the AI looks for. Happy to answer anything about how it works, what it can and can't catch, or why I decided not to store any personal data at all. Comment from Dante Ferraz(@dante_ferraz): Great idea, Mauro. One of the biggest surprises I've seen isn't just hidden fees it's mandatory insurance on rental cars . Many travelers ...

... were individually permitted, correctly permitted, and wrong in aggregate, and an approval layer tuned perfectly still produces your Friday. Approval is a control on discrete acts. Drift is about aggregates, and there is no per call decision that catches it. What catches

... more confident than a month ago. What we shipped? Five integrations now recognize the bugs people actually hit and prove the fix with a real before/after: Stripe, Paddle, Clerk, AgentMail, Descope. Not "we support the API," but "we catch the specific way it breaks." Chaos-order proof for billing webhooks. One of you asked whether we could prove a Paddle subscription survives webhooks arriving in any order. We built it, and then went deeper, to the failure devs actually fear ... ... take on faith. Why we're confident (the part that matters)? We tested recognition against real developer questions, harvested from GitHub/Stack Overflow, phrased by strangers, not us. When recognition was thin, the numbers said so (one integration went from catching

... keep meeting people running one or two service based businesses who reach the same fork in the road. Hire an ops person to catch what is slipping, or find a better way to catch

... Unity 6 pipelines. If you've shipped a game, you know the pain: naming conventions drift, LODs get misconfigured, code reviews miss the same engine-specific mistakes over and over, and performance regressions sneak in until QA catches them. ShintTools catches

... Product Hunt We re building CatchDex, a private collection game for anglers who already take catch photos but want something more useful and fun than a camera roll. The idea is simple: Log a real catch

... fixes they want, preview the repaired email, download the HTML, or continue with visual editing for copy and brand adjustments. The compatibility analysis and repair step runs locally in the browser. Comment from Hossein Yazdi(@hosseinyazdi): Nice addition. Catching those issues before sending can save a lot of time. I also like that users can choose which fixes to apply instead of having the tool automatically rewrite everything. That gives much more control. Comment from Ghost Kitty(@ghost-kitty ...

... Product Hunt We re building CatchDex, a private collection game for anglers who already take catch photos but want something more useful and fun than a camera roll. The idea is simple: Log a real catch

... entire roadmap. Comment from Nolan Vu(@nolan_vu): @thenameisarian  Interesting, alternative to LinkedIn you said "Transparency, not hiding" as the positioning is sharp. The whole industry is moving toward pretending AI involvement isn't happening, so calling it out directly and making it a feature instead of a flaw flips the script in a really smart way. Recruiters needing to know what was AI vs

... free early access and hear what you think Comment from Evgeny Alexeev(@alexe_ev): Here is my 5: Claude - solo building my little product with it! Linear - best jira alternative even for solo Zed - fast and light, if terminal is not enough for you but you won't stick to VS