Christ First | AE @Airtable | Host - MSP Revenue Secrets

Rhondre Giscombe

Disciple of Christ | GTM @Airtable | Building apps & agents for Operators

VP, Head of Product Management @ Airtable

Airtable's MCP server is now live, along with connectors for both Claude and ChatGPT. 

Airtable is increasingly becoming a place where both humans and agents collaborate. By connecting agents to Airtable, you can allow them search and analyze your data, create records and update existing records. 

PMs now frequently ask their agents to synthesize product insights, customer feedback, and roadmap plans stored in Airtable apps -- or even feed this data directly into coding agents to build new features & prototypes! 

We'll be investing heavily in additional MCP capabilities over the coming weeks. Stay tuned for more updates including: 
- Updating Airtable apps via MCP (create new tables and fields)
- Support for users with Interface only access 
- Improved throughput and rate limits 

What additional capabilities would you like to see? Let me know in the comments! 

Try it out: https://lnkd.in/gZiCJtSm

Let me design an Airtable workspace for your company and automate your business processes.

- 📂  I can create a reliable "single source of truth" for your internal data
- 🔀 Map out and automate processes. 
- 🔧 Integrate services you currently use.
- 🌱 I am a fan of flexible solutions that adjust to your needs (AI likes flexible tools as well).
- ✨ I can add generative AI where it can truly help.  
- 👨‍💻 For more complex integrations and automations, I can resort to code 

____

My past experience includes:

- 10 years of web development and building fast web apps with the help of JS and lately Vue.js
- 2 years of business process automation, especially with the help of Airtable, Pipedream
- Being a Co-Founder of Herohero - a successful startup in content monetisation space that is a leading solution in Czech Republic & Slovakia and is expanding to Spain and US
- 1 year of teaching programming in a coding bootcamp
- Being a founder of Otimo - "Save and share the content you truly care about"
- Many years of remote work & collaboration with companies of different sizes & management structures

Martin Malinda

I've been seeing some posts how Airtable is easy to swap with an own vibe coded solution, perhaps to cut costs or to get even more flexibility.

I think this is only a good idea if

- You've actually been using Airtable for a while and you already know you utilize only a small subset of the features
- You're technical and confident to maintain the solution long term (and have someone else to take over if needed)
- Experimenting with AI is part of your expertise / offer, therefore time spent with these tools provides an extra value in terms of learning experience (which you can sell to others)

Having a tiny custom app do only what you need is tempting. Airtable might sometimes seem intimidating with all its different features and you'll save money.


A common success pattern I’ve seen with Airtable is when a company initially just needs a better spreadsheet... meaning a proper table, not a sheet. They don't want formulas to randomly break, they want enforced structure and to link tables, instead of custom v-lookups.

But over time, as they get comfortable, they explore other parts of the product - more advanced formulas, views, rollups, custom lookups, automations, interfaces.

The product does not just "solve a problem", it presents a paradigm, a certain way to think - how to organize and visualize data.

At the moment, even with the most advanced models like Codex 5.3 or Opus 4.6, your vibe coded app starts to degrade as you scale it. Dead code and duplicated code accumulates. You either need to stay small and have a couple of smaller micro apps or you need to be good at architecture and be able to peek under the hood at the actual code and review what's going on there and steer things in the right direction - technically.

I think there's a high chance we're seeing a couple of biases at play. We're seeing the success stories but we don't read about the fail stories - all the cases were people spent countless hours prompting just to get frustrated, not getting what they wanted. Or even abandoning the vibe coded app after some time, seeing that this is not viable enough long-term. Chances are, companies will not present these stories on LinkedIn.

I think it's still perfectly valid to use a well proven solution as a base (pun intended) for your operations. And experiment with vibe coding for niche problems that would be awkward to get working otherwise.

There might be a time when AI coding agents will be able to do good architecture by default and the UX will be top-notch. Where as you prompt, the changes will not be merely *additive*, but the whole app will re-structure as needed. Which is what experienced software software architects and product designers do.

There's already some glimpses of new approaches of multi agent swarms, where multiple agents come together, wearing different hats, to push projects further. But we're not there yet in my opinion, so far every vibe coded project is a code blob, that starts to break apart at some point.

Airtable solutions | Consulting | Business Process Automation | Building Powersave & Powersync.

taplio

The Battle

Rhondre Giscombe

Martin Malinda

Followers

Posts

Engagements

Likes

Performance ScoreRhondre Giscombe's LinkedIn Performance

Performance ScoreMartin Malinda's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Rhondre Giscombe

Martin Malinda

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Rhondre Giscombe's LinkedIn Performance

Performance ScoreMartin Malinda's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreRhondre Giscombe's LinkedIn Performance