Synthetic PM

Company: Synthetic PM

Self-reported basis: The description is entirely from the author’s own submission to a hackathon, unverified and without independent corroboration.

What it appears to be: A tool that uses AI agents to simulate user journeys in software products, aiming to automate product discovery and UI audit tasks for product managers.

What changed: The project was built as a hackathon submission by one person (Bo L) over three days using AI tools like ChatGPT, GPT-5.6-Terra, Browserbase, Playwright, and Notion integration.

Most important open question: Is there sufficient evidence of product-market fit or commercial traction to justify further investment or development beyond a proof-of-concept?

The description states that Synthetic PM explores online software products, reconstructs workflows, and turns the experience into a structured product journey.

• Claimed function: An AI agent signs into any product and maps real user journeys.
• Output format: Structured reports in Notion, with visual screen captures and reasoning traces.
• Technology stack (as declared by author): html, javascript, gpt-5.6-terra, Browserbase, Playwright, ChatGPT.

Inference: The product is a prototype or proof-of-concept built for a hackathon, not a production-ready SaaS offering.

The author claims that Synthetic PM can perform substantial product work—going beyond writing PRDs to actually entering real products and reasoning about them like a product manager.

• Positioning: A tool for product managers to automate UI audits and competitive intelligence.
• Evolution of claim: From a personal problem (maintaining product thinking as companies grow) to a broader idea: whether AI can take on complex, non-routine business work that depends on human judgment.
• Tagline: “An agent that signs into any product and maps the real user journey for product owners, turning an afternoon of manual work into minutes and a few dollars.”

Claim vs. Fact: The author states this is part of a larger idea, but no evidence of adoption or usage beyond the hackathon.

The description indicates that Synthetic PM targets product managers (PMs) who are looking to automate UI audits and competitive intel.

• Target persona: Product owners, especially those using LLMs but seeking deeper workflow understanding.
• ICP inferred: B2B SaaS product teams, particularly those with limited resources or time for manual exploration.

Not evidenced: No customer data, user interviews, or segmentation beyond the author’s personal experience.

There is no explicit mention of pricing, monetization, or business model in the description.

• Author's stated intent: To validate the idea through a hackathon.
• Next steps mentioned: Tune output so reports can be used as input for other agents or decision-making processes.

Inference: The tool may evolve into a SaaS offering, but no pricing or monetization strategy is evident.

The author describes how the product was built using a combination of AI and browser automation tools:

• Tools used: GPT-5.6-Terra, Browserbase, Playwright, ChatGPT.
• Delivery mechanism: Structured screen capture, automated Notion reporting.
• Development speed: Built solo in 3 days.

Not evidenced: No information on scalability, reliability, or robustness of the system beyond a hackathon prototype.

The project is described as a hackathon submission with no evidence of traction or adoption.

• Maturity level: Proof-of-concept.
• User feedback: Limited to the author’s own experience and observation that “seeing the agent work is part of the product.”
• No revenue, customers, or usage data.

Absence of evidence: No signs of real-world use, customer engagement, or product-market fit beyond the hackathon.

The author does not reference any competitors, nor does the description provide context for how Synthetic PM fits into existing tools or markets.

• No mention of competitors.
• Market positioning unclear: The tool is described as a way to automate product thinking but not compared to existing UI audit or competitive intelligence platforms.

Not evidenced: No competitive analysis or market positioning beyond self-description.

Several risks and red flags are evident from the description:

• Single-person development: Team size is 1, which raises questions about scalability and long-term maintenance.
• Prototype-only status: Built for a hackathon with no commercialization plan.
• No monetization strategy: No pricing or business model described.
• Unverified claims: The author states that the tool can “reason like a product manager,” but this is unproven.
• Dependency on AI tools: Reliance on ChatGPT, Browserbase, and GPT-5.6-Terra raises concerns about stability and cost.

Inference: This is a personal project with no commercial viability or traction yet.

1. What specific workflows or product types does Synthetic PM currently support?
2. How does the tool handle edge cases or unexpected behavior in software products?
3. Have you tested the tool with actual product managers or teams? If so, what feedback did you get?
4. What is your plan for monetization and scaling beyond the hackathon prototype?
5. Are there any legal or ethical concerns around AI agents accessing and interacting with third-party software?