FixMyWorkflow

Note

All claims are based on the author's own description and are unverified. No evidence of revenue, customers, traction, or actual product usage exists beyond what is stated.

The description states that FixMyWorkflow enables users to describe a business process in plain English. It then uses AI (specifically OpenAI) to generate structured workflow components such as intake forms, ordered steps, task templates, and owner roles. These are intended to be activated when a user submits a request, creating a live workflow run with progress tracking.

The system is built using Phoenix, LiveView, Ecto, PostgreSQL, and integrates with the OpenAI API. It supports persistence of workflows, intake forms, runs, tasks, and related data, all isolated by organization.

It also includes a fallback mechanism for when AI responses are unavailable — a deterministic demo sample that mimics the full workflow path without relying on live AI calls.

Inference The product appears to be a proof-of-concept or prototype, not a production-ready SaaS offering. It is described as having been built during a hackathon event.

The author claims FixMyWorkflow turns messy business processes into working automations through one conversation — distinguishing it from tools like ChatGPT that only provide instructions.

Early versions focused on generating blueprints, but the author notes a key pivot: the useful moment comes not from the blueprint itself, but from its activation and ability to receive real requests and track work.

This evolution suggests a move from idea generation toward execution and usability — emphasizing functionality over novelty.

Claim

The tool is positioned as an AI-powered workflow automation platform that moves beyond static advice into actionable workflows.

Inference This positioning implies a shift from a generative AI assistant to a workflow engine, though no evidence supports whether this transition has been realized in practice.

The description does not explicitly define the target customer or ideal customer profile (ICP). However, it references small businesses and teams managing property maintenance processes. The demo focuses on tenant repair requests, suggesting a use case for small business owners or facility managers who want to automate repetitive tasks.

Claim

The tool targets users with informal or unstructured workflows that need automation.

Inference Based on the example, the ICP likely includes small teams or individuals managing simple but recurring processes — not enterprise-level organizations.

There is no evidence provided about pricing, monetization, or business model. The author describes a demo and development process but does not mention any revenue streams, subscriptions, or paid features.

Not evidenced

The system uses:

• Phoenix (web framework)
• LiveView (real-time UI updates)
• Ecto (database modeling)
• PostgreSQL (data persistence)
• OpenAI API integration
• Codex with GPT-5.6 for development assistance

It includes transactional persistence, validation of AI outputs, error handling, and a fallback demo path.

Inference The tech stack suggests a modern web application built for interactivity and data consistency. However, no evidence indicates deployment, scalability, or infrastructure beyond the hackathon prototype.

There is no evidence of traction, customers, revenue, or adoption. The project is described as a hackathon submission with no mention of users, usage metrics, or market validation.

Not evidenced

The author does not reference competitors or the broader marketplace. The only comparison made is to ChatGPT — implying that FixMyWorkflow offers something more than generic advice.

Not evidenced

• Unproven execution: The product is described as a demo-only experience, with no indication it has moved beyond prototype or been tested in real-world conditions.
• AI dependency: Reliance on OpenAI API for core functionality introduces risk if the service becomes unavailable or changes.
• Limited scope: No evidence of integrations, advanced features, or scalability beyond the single demo use case.
• No commercial viability: No mention of monetization, pricing, or business model.

Inference The tool may be a compelling idea but lacks evidence of real-world utility or commercial potential.

1. Has the workflow generation been tested with multiple users or processes beyond the demo?
2. What happens if the OpenAI API fails during a live workflow run?
3. Are there plans to support integrations with existing tools (e.g., Slack, email, CRM)?
4. How is data isolation handled across organizations?
5. Is there any plan for editable blueprints or user customization of generated workflows?
6. What are the technical limitations of the current architecture that would prevent scaling?