Show HN: Mediator.ai – Using Nash bargaining and LLMs to systematize fairness
Nash bargaining + LLMs = fair deal automation. Mediator.ai just launched to systematize what mediators charge $5k+ for.
Why it matters
An indie founder is shipping an LLM-powered product that applies game theory (Nash bargaining solution) to real-world negotiation. It's a novel app-layer use of LLMs for a high-friction, high-value problem (legal/financial mediation), but early-stage and self-launched with minimal traction.
The key facts
11 to knowUses LLMs for preference comparison to estimate utility functions (solving Nash's 1950s limitation)
Genetic algorithm optimizes agreement terms across multiple parties
Soft-launched over weekend with 7 HN points, 0 comments
Targets high-value negotiation scenarios (prenups, settlements, multi-party deals)
Unverified early traction — no user numbers, revenue, or adoption data provided
Soft launch: April 20, 2026
Core innovation: LLMs used for preference comparison (not direct utility estimation) feeding into genetic algorithm optimization
Use case: Multi-party negotiation mediation (prenups, divorces, business disputes)
Technical foundation: Nash bargaining solution from 1950s game theory + modern LLM capability
Founder motivation: Eight-year personal problem (prenup mediation) as seed for product
Currently early-stage: 7 HN points, 0 comments at time of soft launch
Go to the source
Hacker Newsmediator.ai
Publisher excerpt: Eight years ago, my then-fiancée and I decided to get a prenup, so we hired a local mediator. The meetings were useful, but I felt there was no systematic process to produce a final agreement. So I started to think about this problem, and after a bit of research, I discovered the Nash bargaining…