TLDR; Mohammad provides the most insane breakdown all the no code apps in his stack. He demonstrates how he integrated with Twitter. Fascinating to see below…
1. What did you make?
MN: I built Twitter Argumeter, a tool that lets you find out how much a Twitter user likes to argue and creates an ArguMeter profile of the user by doing an analysis of the percentage of tweets that are comments and by doing an average sentiment analysis of those tweets. It also shows you the most positive and the most negative tweet by the user ever.
2. What is the most interesting no code stack that you made.
MN: Twitter Argumeter for sure but Works For All – Meetup Place Finder is a close second.
3. What is your most favorite thing that you made.
MN: Capria VentureBasecamp where I am the Product Manager. I ideated the product with the co-founder when we started and then took 2 months to build the whole online product on my own using Bubble.io. It uses 45+ years of VC investing experience to support entrepreneurs in incubators to become investment-ready. Although the product is B2B, we have an open investment readiness assessment on our website (built on no-code using Typeform, Airtable, and Zapier) here to provide a sneak-peek of the product.
4. Is Twitter API possible to use without code?
MN: Yes. The only code you need to understand is how to make API calls on the No-code tool you are using. Some tools like Bubble have plugins to make Twitter API calls. So, you don’t even have to understand how to set up an API. Just apply for a developer account with Twitter and then enter your Twitter API token in the plugin.
5. What problems were you trying to solve, your motivation, what is your why for this?
MN: The idea to build Twitter Argumeter cropped out of a tweet by a16z’s Andrew Chen. He wanted Twitter to put a flag of “this person likes to argue on the internet” on user profiles. This stems from the genuine frustration of Twitter users over trolls. Many times, it becomes hard to have a constructive debate with these users and before you realize it, you have gone down the rabbit hole of defending yourself, your character, and your values. So, it is better, especially for influencers, to know the twitter trolling or argument history of such users before engaging with them online or even in-person.
6. What different tools did you use?
7. Can you talk specifically about why you chose those tools? And how did you figure out to use those tools? Were there any others you considered or failed with at first?
MN: My favorite no-code tool is Bubble and it would have been much easier to implement this workflow there. However, I stayed away because of a few reasons. Bubble needs a lot of effort in designing a good UI. Also, its free plan is very slow in processing, especially when you have API calls set up. There is also an issue with making API calls process synchronously after a workflow action is complete. API calls are called asynchronously and prematurely before the input data is even ready. As my app required a minimalistic UI, a bunch of API calls and backend processing, I chose to go with Glide which provides a beautiful mobile app type UI right away to work on. Also, as it connects to Google Sheets for the database, I had the freedom to use any workflow automation software under the sun to do the Twitter data processing.
Next, I was looking for the right tool to process my Twitter data. I played around with Standard Libray which provides Zapier-like trigger-leads-to-action workflows but its Builder UX was a big roadblock. Unlike Zapier, you can’t design what each trigger and action does until you have finalized the workflow which can’t be edited later unless you switch to the code editor. Generally, when I am designing a workflow, I end up testing a few iterations to get the workflow right. Also, I needed to clean up, parse and process the data before it was sent back for consumption by the user-facing Glide app. I found the perfect solution in Parabola. First of all, they have a great builder UX and the inline documentation for each element is super helpful. I was able to parse the data received from Twitter API and process it to create the Argumeter profiles, all in Parabola, especially with the help of their in-built Sentiment Analysis element.
Finally, I decided to use Zapier to trigger my Parabola flow through a webhook when the user entered a Twitter username on Glide which created a new Google Sheet row with the username. Parabola webhook is in beta and you have ask their support to activate it. They were quick to respond and happy to help in creating the flow if you explained the use case. True customer success. The resultant Argumeter profile ended on Google Sheets adding a new row of data for Glide to consume and display.
8. What would you do differently building it or something valuable you learned you’d like to share with other makers?
MN: If I had to build Twitter Argumeter again, I would rebuild it in Bubble to reduce the time it takes to generate a profile as the free Zapier account takes about 15 minutes to trigger the workflow. That wait time is too much for a quick fun app that you check out at a small work break.
9. Please provide what links to your project website, your twitter, where can people return the love?
Upvote it on Product Hunt.
If you like my work, you can buy me a chai here
10. Is there anything else you would like to share or some feedback/request/action that you’d like to ask the Side Project Stack Audience to do for you?
MN: Do try out my other products on Product Hunt and share your feedback & suggestions in comments. My Product Hunt profile is here.
Also, sign up for my upcoming social commerce app Swagbook on Product Hunt.