I think I'm going to be writing these on the day I get back from a weekend. I've realized that I'm too burned out by Friday to write a whole dev blog. Plus, my Friday standup notes don't actually include what I complete on Fridays, whereas my Monday notes too. Therefore, I think it's a logical choice.
This was a long weekend, hence why I'm posting on Tuesday.
Last week began with a determined effort to address some technical hiccups that were detracting from user satisfaction. On Monday, the 12th of February, I tackled two significant issues. The first involved fixing a malfunctioning cancellation button that was not appearing for trial cancellations, a critical feature for user autonomy and trust. Concurrently, I worked on a glitch with the onboarding chat buttons, which were getting stuck and impeding the user's progress through the app.
The following day, on the 13th, my efforts to refine the onboarding experience continued as I fixed that issue that I was working on with the onboarding. Additionally, I reinstated a table for AppSumo, essential for deploying new code to the production server without hitches. For some reason, it keeps getting deleted by itself periodically. Changes to the welcome email content, as specified in our Notion documentation, were also implemented, ensuring new users received the intended initial communication.
Midweek, on Valentine's Day, I addressed a new challenge: creating an error message for when a request times out. This was particularly necessary after our access to the OpenAI API was temporarily lost due to an unpaid bill, affecting our backend's ability to send bot messages promptly.
By the 15th, my focus shifted back to refining the cancellation functionality and correcting minor typos scattered throughout our platform. The inconsistent use of English by the offshore team had led to a series of small, yet bothersome, errors, including inconsistent capitalization. However, the development of the cancellation button was momentarily stalled by an unexpected bug related to Stripe API calls inadvertently reverting plan statuses in our database. I worked late into the night to try to get this done as soon as possible. With it, I also placed a little text under the cancelation button to direct any users to the help email address.
The week's efforts culminated on Friday, the 16th, with the successful resolution of the cancellation functionality. The breakthrough came upon realizing that a webhook, previously overlooked because my code was running locally connected to the staging server database, was the culprit. This setup was necessary as I couldn't get Stripe to function locally. Additionally, I managed to deploy minor copy changes and adjustments to the action bar width that had been pending. Over the weekend, I had to revert the timeout error solution introduced on Wednesday, as it was erroneously timing out active responses. Thankfully, our access to the OpenAI API was restored on Friday after the CEO addressed the outstanding bill.
This sequence of events underscores not just the technical challenges faced in software development, but also the importance of persistence, attention to detail, and the occasional necessity to step back, reassess, and tackle a problem from a new angle. The week was a testament to the ongoing journey of improving our application, ensuring that each user's experience is as smooth and enjoyable as possible.
Hopefully we can secure our funding as soon as possible; I'm really looking forward to getting more people on our team. Until next time, cheers.
Top comments (0)