I’m a solo dev building a tool to replace the "pros and cons" list for high-stakes decisions.
While LLMs are great for brainstorming, I found them dangerous for actual decision-making—they hallucinate and try to please you rather than giving you a hard truth. I wanted something that combined deterministic logic (weighted criteria) with generative explanation (narrative).
The Product: It’s called Decision Audit. The first module is "Should I Quit?" (career transition analysis). Instead of a chat interface, it uses a structured interview to calculate specific indices (Burnout, Runway, Decision Rights, Optionality). The output isn't a text blob; it's a 3-page structured Dossier (PDF) that includes:
An "Exit Lean" score (calculated via weighted algorithm, not LLM sentiment).
A "Risk Matrix" (flagging specific dangers like visa dependency or low runway).
A 30-day "Validation Sprint" (action plan).
The "Brain": A rules-based scoring engine (JSON logic) handles the math. I only use the LLM API at the very end to summarize the data points into the narrative sections of the PDF.
I’d love feedback on the scoring weighting—specifically if the "Chaos Index" feels too sensitive.