Feedback on No-code Development in the AI Era & Suggestions for AppSheet

[hien_nguyen]

Dear AppSheet Development Team,

Over the past 7–10 years, many users—especially non-developers—have turned to no-code/low-code platforms like AppSheet because of their accessible drag-and-drop approach, enabling them to build functional apps without deep technical knowledge. Despite limitations in customization or UI/UX polish, the core value was speed and ease of use.

However, we are now entering an era where AI is rapidly transforming how software is built. Prompt-based AI tools can now generate fully functional code and workflows with impressive speed and accuracy—even by users with no IT background. For many tasks, AI has become a viable (and sometimes superior) alternative, even replacing roles traditionally filled by junior developers.

This brings up a serious question: What is the future of no-code platforms like AppSheet in this AI-driven landscape?

From my experience as a no-code solution developer, I’ve noticed that:

• AI integration in AppSheet is still quite limited.
• Users still spend a lot of time manually setting up data models, views, actions, and formulas.
• Tasks like OCR, intelligent form parsing, or working with large datasets are either not yet fully supported or lack the precision needed in real-world business scenarios.
• Chart Gallery and UX too poors
• In contrast, code-based platforms (augmented by AI coding assistants) can auto-generate many of these functions with better flexibility.

As a result, building with no-code is no longer always faster or more efficient than coding with AI.

That said, I strongly believe AppSheet still has great potential—especially for enterprise use and teams without strong technical backgrounds. But to stay competitive and relevant, AppSheet must evolve. Here are a few suggestions:

1. Deeper integration with AI tools like Gemini or GPT:

• Auto-generate views, actions, and expressions from plain language prompts.
• Offer contextual AI suggestions based on app usage and data patterns.

2. Smarter automation for common workflows:

Auto-detect approval flows, task assignments, and update actions based on column names and usage patterns.

3. Improved AI capabilities in OCR, document analysis, and predictive modeling:

Leverage Google’s strong AI/ML stack to enhance features like AppSheet OCR, AI predictions, etc.

4. Focus on enhancing enterprise features:

Role-based dashboards, real-time performance analytics, and integrations with BigQuery or Google Cloud APIs.

No-code is not just about avoiding code—it’s about accelerating solutions. AI can and should be a powerful co-pilot in this journey.

As a dedicated user and advocate of AppSheet in the digital transformation space, I hope to see more rapid innovation that bridges the power of no-code with the intelligence of AI.

Looking forward to seeing AppSheet thrive in this new era.

Best regards,

Hien Nguyen

[hien_nguyen]

Technology is evolving rapidly, and creating an app with AI-powered code has become much more accessible to many people. Meanwhile, AppSheet hasn’t seen many feature updates, and it's plagued with numerous bugs. Not to mention the lack of AI integration—many essential features that users really need are not being prioritized or improved. I believe if AppSheet doesn’t make significant changes soon, users will likely abandon it for other platforms that better suit their needs or choose to build coded applications instead, where they have more freedom and flexibility with AI-assisted development.

[Rifad]

What do you mean by AI integrations? I wouldn't recommend that AppSheet include native AI integrations, as this is just a passing trend. Most needs can be effectively addressed using custom and third-party tools. This may involve complex setups like webhooks, document parsing, flow controls, and data transformation. Think of AppSheet as a no-code front end rather than a back end, where various complexities can be handled with different tools already.

For AI-assisted development, I don’t think it's necessary for a developer who is skilled 6/10 in AppSheet. They will likely be faster at clicking, navigating, and writing expressions than trying to communicate with AI, which may generate incorrect code. It's a straightforward task, and they can quickly move on to the next table or expression. Using AI in this context can hinder productivity (I primarily use AI for workflow automation, not for AppSheet related coding).

---

@hien_nguyen wrote:

Tasks like OCR, intelligent form parsing, or working with large datasets are either not yet fully supported or lack the precision needed in real-world business scenarios.

---

You may use something like this: https://www.llamaindex.ai/llamaparse I wouldnt recommend document OCR also in Appsheet as it can lead to problems and OCR can vary depending on usecase. When there is already excellent backend tools I think appsheet needs to focus on its niche and trajectory and not unwanted tools like this.

---

@hien_nguyen wrote:

Users still spend a lot of time manually setting up data models, views, actions, and formulas.

---

This might be good for most basic least complex setups. Maybe its just 10% of the total developement. Rest of the business logic is supposed to be develoepd by a developer.

---

@hien_nguyen wrote:

Chart Gallery and UX too poors

---

Agreed to that

---

@hien_nguyen wrote:

1. Deeper integration with AI tools like Gemini or GPT:

• Auto-generate views, actions, and expressions from plain language prompts.

---

I would refuse to do this for any serious enterprise clients. For personal usecase and quick prototype yes 100% useful

In conclusion, I wouldn't recommend native AI integrations due to the broad scope of business needs. Instead, AppSheet could focus on improving features like faster webhooks, streaming, or WebSockets for quicker, real-time responses.