Show HN: I built developer tooling for the Airtable API that I needed (airwalker.io)

73 points by goksan ↗ HN
As a software engineer, I've experienced firsthand the challenges of working with the Airtable API.

As more non-technical users began using the platform in our growing business, the need for engineering to automate processes and sync data into Airtable grew. However, keeping track of process failures and ensuring that no unresolved failures slipped by was difficult and required significant effort.

That's why I created Airwalker, a toolkit that improves the reliability of processes using the Airtable API and helps you correct issues quickly and with minimal effort.

Here are some of the features Airwalker offers:

* Base schema timeline

* Request/response logging

* Edit & replay

* Custom automatically maintained TypeScript types

Airwalker is free to use right now, and I welcome any feedback or suggestions.

29 comments

[ 0.27 ms ] story [ 72.0 ms ] thread
Airtable is amazing to use as a kind of combination of a Database and product Admin panel.

The API is indeed a little bit clunky, but it works fine.

I've had to write my own Airtable handler (e.g.: API response schema validation and type translation) codes for all my projects where I've used it, maybe this could actually be a great unified solution.

Edit: The Superinverse logo is nice.

Hey, thanks for sharing

I've also thought about additional an additional layer of validation & type transformation, one for roadmap.

I like the Superinverse logo too!

Oh this is nice! I've written my own wrapper "Airfetch" around it to just... work, but it's so dirty I can't actually share it (ha) but it works super well, and I use Airtable for everything.

Thanks so much for sharing this!

Hah we both had the idea of prefixing it with Air

Thank you for sharing, feel free to reach out if I can help out.

I've been using Airtable to run Demuxed (decently sized video conference for people that work in video[1]) since ~2016 and my interest is definitely piqued. We do talk submission, anonymized selection processes, reviews, sponsor management, everything. We've built some custom UIs around things like the talk review process and the API is workable, but things like types would have been a god send.

I was a little taken aback when I went to sign up for the free trial and got hit by a Stripe checkout, though. Yes, checkout was for £0.00, but still, it was enough to make me close the tab at that point. This project is interesting but probably still a long shot for our workflows, so that (admittedly fairly low) barrier was a blocker.

Best of luck!

[1] https://2022.demuxed.com (last year's conference site)

Interesting to hear how you're using Airtable, thanks for sharing.

Happy to talk more about your workflows if you'd like, but I appreciate you might already be able to tell whether Airwalker would be a fit or not.

I would love to learn more about your workflows, especially steps that end up requiring custom solutions, like UI that you mentioned. Care to share?
My biggest gripes with the Airtable API for a small project that has a couple of tables with > 100 rows were around pagination and joining the data between the tables which was super slow.
Thanks for sharing, I've heard similar gripes before. Airwalker doesn't do anything to ease that pain yet, one for consideration
My company has an Airtable competitor that has API support for SQL. It has pretty good SQL capabilities solving the issues with pagination you have in Airtable.

An overview of the supported SQL syntax is here:

https://blaze.today/datablaze/docs/bsql/

No joins yet between tables yet, but they are planned. (You can use lookup fields to achieve them to some extent, but that requires planning ahead and is limited).

Congrats for shipping!

I tried to open your Terms and Privacy Policy pages but both are spitting a 400 error after timing out.

Thanks so much for letting me know about this.

I'm not able to reproduce this currently, maybe it's fine on your end now too?

Indeed, it works fine now. I'm curious, what did you use to build your website?
It's built using Go and html/template from the standard library.

For interactivity I wrote some custom JS, I think it's similar to libraries like https://htmx.org or https://turbo.hotwired.dev, this stuff is inside the application though.

Let me know if there's anything you were particularly interested in knowing.

Love the simplicity of the stack!

To be honest my curiosity was more of an engineering itch, what kind of setup would exhibit the failure mode I observed (a specific static route timing out).

Have you considered rendering HTML pages statically to avoid situations like this?

(By the way don’t let this conversation distract you from focusing on your product, your website is already excellent! And I appreciate you taking the time to reply :)

Hey no worries, happy to share :)

I should mention that the terms are fetched via an external API, I think that API failing temporarily caused the issue you were observing.

You’re right, static generation would have avoided the issue. Alternatively, I could be caching the response for a period. I’ve also got HAProxy in front of the application servers, but haven’t explored it’s caching capabilities.

I am glad to see an example of a better-built API which has pricing terms attached. For someone working in a corporate setting who needs a great off-the-shelf API to get an integration working ASAP, it seems to be just at the threshold of needing a purchase order to get it.

For 'goksan, I'm curious what kind of research you did on the customer base for this, and how you thought through pricing? It feels attractive to me, but curious if there's more rigor involved there.

Hey, thanks for commenting.

The API still a subset of the Airtable API for the moment to avoid requiring rewrites of existing integrations. I do plan to expand the API in the future to enable novel features.

Most of my research is in talking to other engineers using Airtable, admittedly none used Airtable as widely as we did at my last job.

I want to deliver the best possible experience for Airtable API integrations, and I believe pricing in the speculated range is necessary to sustain that focus. It's something I'm still thinking about.

Moving out of Airtable and replicating all of it's goodies is a tricky and expensive task, I'd like to help customers avoid that where possible, or delay it at the very least.

Love the idea! Have experienced some of the pain points of working with Airtable myself. Keep building :)
A name that I couldn't forget!

Hope everything is going great out there :)

Airtable is great for end users and internal tooling
This is pretty cool, thanks for sharing
Thanks for the kinds words :)
hey I love the work that you are doing here, I have been wanting to create dev oriented data table service, like AirTable but more dev friendly with a strong plugin system. Want to have a quick chat?
let's do it, email and twitter in HN bio
Have you thought of repacking your Edit&Replay capability into a stand-alone tool able to edit-and-replay any request? I've been at a number of enterprises that have been such a tool homegrown but with limited capabilities?
Hey, thanks for sharing your thoughts.

I haven't thought about it too much personally, I think it'd be super useful though!