This open-source RL model scores a sales chat after every message.
This is a AI post classified by Jev as AI sales & support (a tool drop), kept by the AI Radar because it carries real work, not commentary.
This open-source RL model scores a sales chat after every message. Not “write a better reply.” It outputs: close probability just jumped 78%. - Trained on 100k+ synthetic chats. - can Fine-tune on your own transcripts. - PPO + embeddings - 34.7% better than LLM-only - 96.7% conversion-prediction accuracy - 85ms inference vs 3450ms for GPT-4 - +43.2% conversions when reps used the live score - MIT, 385 downloads
Posted by Md Ismail Šojal 🕷️ (55.9k followers) 1 h ago · 3 likes · 730 views · view the original post on X. Kept by the AI Radar as AI sales & support.
More AI work like this
- Ever wonder how AI can boost sales? This reinforcement learning model analyzes… — @HuggingModels
- Decagon is growing across Belgium, the Netherlands, and Luxembourg. 🇧🇪🇳🇱🇱🇺 — @DecagonAI
- Here's a clip from @ENowoslawski. He's already using Grok Bot + Jev to build on @Clay… — @clay
- "Training an AI SDR to do outbound or inbound really isn't that hard. — @jasonlk
- Our old contact form took a day to get a response. — @saastr
- Codex + computer use + Linkedin — @cccyd_qwq
- Verdigris left Salesforce after a decade to build an AI-native GTM on Lightfield. — @lightfld
- Access Futurum intelligence in the way that works best for your team. — @TheFuturumGroup
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 40.6k posts from 5k X accounts over the last 14 days, 1.7k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-22 00:34 UTC. Full method.