Articles on shipping real product.

The Bubble Performance Ceiling: When to Migrate Your App to Code

Bubble's performance ceiling shows up in five predictable ways. Here's how to recognise the signals, measure them honestly, and decide whether to refactor, extend, or migrate.

28 May 2026Read →

12 min read

Bubble to Next.js: A Step-by-Step Migration Guide for Founders

How to migrate a Bubble app to Next.js without big-bang rewrites. The target architecture, the six phases, the gotchas, and how to keep shipping while you migrate.

22 May 2026Read →

11 min read

Migrating Bubble Data to Postgres: The Complete Reference

How to move every Data Type from Bubble to Postgres without corrupting data. Idempotent imports, soft-delete handling, option sets, recursive foreign keys, reconciliation.

15 May 2026Read →

Bubble vs Code: The Real Cost at 10k, 100k, and 1M Users

Honest cost comparison between Bubble and a code stack (Next.js + Supabase + Vercel) at three scale milestones. Infra, developer time, hiring, and total cost of ownership.

8 May 2026Read →

Why External Backends (Xano, Supabase, Firebase) Kill Bubble Apps

Bolting an external backend onto Bubble looks cheap. The hidden costs - latency, sync bugs, data drift, double maintenance - quietly destroy the app. Here's what to do instead.

30 Apr 2026Read →

Bubble API Workflows to Trigger.dev: How to Migrate Async Work

Mapping Bubble's API workflows, scheduled workflows, and recursive workflows onto Trigger.dev tasks. The patterns, the gotchas, and what you gain.

22 Apr 2026Read →

11 min read

How to Audit Your Bubble App Before You Commit to a Migration

A 50-item checklist for auditing a Bubble app before migration. Covers data model, workflows, integrations, plugins, performance, and the AMBIGUOUS items that surface as risks.

12 Apr 2026Read →

Slice-by-Slice vs Big-Bang: Choosing Your Bubble Migration Strategy

The two migration strategies, how to choose, and why slice-by-slice almost always wins. Migration risk concentration, business continuity, and the rare cases big-bang is right.

4 Apr 2026Read →

Translating Bubble Privacy Rules to Postgres RLS Policies

How Bubble privacy rules map to Postgres row-level security. Common patterns, gotchas where Bubble's rules silently fail, and a worked example with Supabase.

26 Mar 2026Read →

See the product development service →

The Hidden Costs of Staying on Bubble After You've Outgrown It

The visible cost of staying on Bubble is the monthly bill. The hidden costs - lost hires, slower shipping, vendor risk, valuation drag - add up to much more.

18 Mar 2026Read →

AI Product Development

For founders and teams building AI products - architecture, the stack, evaluation, cost, UX, and getting from prototype to production.

How to Build an AI MVP in 6 Weeks (Without Cutting Corners That Matter)

A realistic 6-week plan for shipping an AI product MVP: what to build, what to defer, and the corners you can cut versus the ones that sink you. By an ex-Airdev senior engineer.

1 Jun 2026Read →

AI-Native Architecture: Designing Products with AI at the Core

What it means to build AI-native instead of bolting AI on. The orchestration layer, the data layer, observability, and the design decisions that make AI a first-class capability.

27 May 2026Read →

The AI Product Development Stack: What We Use and Why

The stack we reach for building AI products - Next.js, Supabase, Trigger.dev, OpenRouter, and more - with the reasoning behind each choice and when we deviate.

23 May 2026Read →

RAG vs Fine-Tuning vs Prompting: Which Approach for Your AI Product

A decision framework for the three ways to make an AI model do what you want: prompting, retrieval-augmented generation, and fine-tuning. When each fits, and when to combine them.

19 May 2026Read →

How to Validate an AI Product Idea Before You Build It

AI products fail for the same reasons all products fail, plus a few AI-specific ones. How to validate demand, feasibility, and unit economics before writing code.

15 May 2026Read →

Building AI Features That Don't Hallucinate: Guardrails and Grounding

Practical techniques to stop an AI product from making things up: grounding in real data, output validation, refusal behaviour, and the guardrails that keep AI trustworthy.

11 May 2026Read →

The Cost of Running an AI Product: Token Economics Explained

How AI products actually cost money: tokens, context windows, model tiers, and the levers that control your inference bill. Plus how to model unit economics before you scale.

7 May 2026Read →

Streaming AI Responses: Building Responsive AI UX

Why streaming is essential for AI product UX, how it works technically, and the patterns that make AI features feel fast even when generation takes seconds.

3 May 2026Read →

Evaluating AI Output: How to Test a Non-Deterministic Product

You can't unit-test an LLM the way you test deterministic code. How to build evaluation sets, score non-deterministic output, and catch regressions before they ship.

29 Apr 2026Read →

AI Agents vs Workflows: When to Use Which

Agents are flexible but unpredictable; workflows are reliable but rigid. A practical framework for choosing between agentic and workflow approaches in your AI product.

25 Apr 2026Read →

Designing the Human-in-the-Loop: AI UX Patterns That Work

The UX patterns that make AI products trustworthy: review-and-approve, suggest-don't-act, confidence signals, and easy correction. How to design the human-AI handoff.

21 Apr 2026Read →

Vector Databases Explained: pgvector vs Pinecone vs Others

What vector databases do, why AI products need them, and how to choose between pgvector, Pinecone, and dedicated options. Plus when you don't need a separate one at all.

17 Apr 2026Read →

Prompt Engineering for Production: Beyond the Playground

Prompts that work in the playground break in production. How to write, version, test, and maintain prompts as real software artifacts in an AI product.

13 Apr 2026Read →

How to Add AI to an Existing Product Without a Rewrite

You don't need to rebuild your product to ship AI features. How to add AI as a bounded service alongside your existing app - whatever stack it's on.

9 Apr 2026Read →

Multi-Model Architecture: Why You Should Never Hardcode One LLM

Hardcoding one model provider is a liability in a market that changes monthly. How to build a model-agnostic architecture with a routing layer, and what it buys you.

5 Apr 2026Read →

Building AI Products on Supabase: Auth, Vectors, and Edge Functions

Why Supabase is a strong foundation for AI products: Postgres with pgvector, row-level security that extends to retrieval, auth, and edge functions - all in one place.

1 Apr 2026Read →

AI Product Observability: Logging, Tracing, and Debugging LLM Calls

When your AI does something weird in production, you need to see what happened. How to trace LLM calls, what to log, and how to debug a non-deterministic system.

28 Mar 2026Read →

Shipping AI Safely: Rate Limits, Fallbacks, and Graceful Degradation

The operational safety net for AI products: rate limits, provider fallbacks, graceful degradation, prompt-injection defence, and cost controls. What to build before you launch.

24 Mar 2026Read →

From Prototype to Production: Hardening an AI Demo for Real Users

The AI demo that wowed everyone is 20% of the work. The other 80% - reliability, edge cases, evaluation, observability, cost - is what makes it a product. A hardening checklist.

20 Mar 2026Read →

See the AI transformation service →

AI Product Pricing: How to Charge When Your Costs Are Variable

Traditional SaaS pricing assumes near-zero marginal cost. AI products have real per-use costs. How to price so you don't lose money on power users - seat, usage, and hybrid models.

16 Mar 2026Read →

AI Transformation

How traditional businesses turn their data and operations into real, shipped AI products - by industry and by tool.

AI for Professional Services Firms: Where to Actually Start

Where law, accounting, and consulting firms get real ROI from AI. The high-impact use cases, the ones to avoid, and how to ship without risking client confidentiality.

30 May 2026Read →

AI for E-commerce: Beyond the Chatbot

The AI use cases that actually move revenue for e-commerce and retail businesses: personalised search, returns automation, recommendations grounded in real inventory, and support deflection.

20 May 2026Read →

AI for Real Estate and Property Management

Where AI creates real value in real estate and property: listing automation, tenant communications, lease and document extraction, and internal ops tools. With a marketplace example.

12 May 2026Read →

AI for Financial Services: Compliance-Safe Automation

Where fintech, lenders, and financial advisors get value from AI without falling foul of compliance: KYC document processing, risk triage, internal research assistants, and audit-ready design.

4 May 2026Read →

AI for Healthcare and Wellness Businesses

Privacy-first AI for clinics, wellness businesses, and health platforms: intake automation, scheduling assistants, documentation support, and internal tools with strict data boundaries.

26 Apr 2026Read →

Building an AI Chatbot That Actually Knows Your Business Data

How retrieval-augmented generation (RAG) lets an AI chatbot answer from your real data with citations - and why most business chatbots fail without it. The architecture, explained.

18 Apr 2026Read →

AI Document Extraction: Turning PDFs and Invoices into Structured Data

How AI turns unstructured documents - invoices, contracts, forms, statements - into clean structured data. The architecture, accuracy techniques, and where it pays back.

10 Apr 2026Read →

AI Workflow Automation: Replacing Manual Work, Not People

How to automate operational workflows with AI without the brittleness of pure rules or the risk of full autonomy. The patterns, the human-in-the-loop design, and where to start.

2 Apr 2026Read →

Choosing an AI Model in 2026: Claude vs GPT vs Open Models

How to choose between Claude, GPT, and open-weight models for a business AI feature - and why the right answer is to stay model-agnostic via a routing layer like OpenRouter.

25 Mar 2026Read →

See the Canvas rescue service →

The AI Transformation Audit: What a Good One Actually Delivers

What an AI transformation audit should produce: an opportunity map scored by impact and effort, a costed roadmap, and an honest verdict - including when AI is not the answer yet.

16 Mar 2026Read →

Canvas (Airdev)

Reference material for owners and operators of Canvas (Airdev) Bubble apps - maintenance, debt audits, upgrades, and the long-term migration path.

Canvas v5 to v6: Should You Upgrade, Rebuild, or Migrate?

Canvas v5 apps can't upgrade in-place to v6. Here's the honest framework for choosing between rebuilding on v6, staying on v5 with maintenance, or migrating off Canvas entirely.

25 May 2026Read →

The Canvas Tech Debt Audit: 20 Things to Check Before Adding Features

Twenty concrete checks for technical debt in a Canvas (Airdev) Bubble app. Run this audit quarterly and before any major feature push to avoid compounding debt.

18 May 2026Read →

A Quarterly Maintenance Checklist for Canvas Apps

The 15-item maintenance checklist we recommend running quarterly on every Canvas (Airdev) Bubble app. Catch issues before they become incidents.

10 May 2026Read →

Canvas Reusable Elements: Best Practices and Anti-Patterns

Reusable elements are Canvas's biggest leverage and biggest debt source. The patterns that work, the ones that don't, and how to refactor an app full of bad reusables.

2 May 2026Read →

Canvas Privacy Rules: How to Audit What Your App Exposes

A step-by-step audit of Canvas Bubble privacy rules. The common ways data leaks via the Bubble Data API, how to test for it, and how to lock down what should be private.

24 Apr 2026Read →

Canvas Performance: Speeding Up Searches and Page Loads

Practical techniques for making Canvas Bubble apps faster. Database search optimisation, page weight reduction, repeating group patterns, and when to denormalise.

15 Apr 2026Read →

Canvas Plugin Strategy: What to Keep, Replace, or Remove

How to audit, prune, and choose plugins in a Canvas Bubble app. The hidden costs of plugin sprawl and a framework for deciding which plugins earn their place.

7 Apr 2026Read →

Working With (and Around) the Canvas Style System

The Canvas style system is powerful but has real limits. The patterns that work, where it breaks, and how to keep design consistency on a Canvas app over years.

30 Mar 2026Read →

Adding AI Features to a Canvas App Without a Rewrite

How to add Claude, GPT, and other AI features to a Canvas Bubble app without leaving the framework. Patterns that work, patterns that do not, and when the AI work itself signals you should migrate.

22 Mar 2026Read →