Show HN: GPT-5.1 reasoning agents for hypertrophy periodization

I’ve been lifting for years using complex methodologies (Mentzer HIT, Kuba Method, FST-7) and got frustrated that existing trackers are just glorified spreadsheets. They record what you did, but they don't tell you what to do based on your fatigue or biomechanics.

So I built Arvo (https://arvo.guru). It’s a "context-aware" AI coach, not just a logger.

The Problem: Standard apps use static linear progression (add 2.5kg every week). But if you are following a fatigue-based system (like Mike Israetel’s RP or Kuba), you need to autoregulate based on RPE/RIR and soreness. Most users fail to track this accurately in Excel.

The Solution & Tech Stack: Instead of if-then statements, Arvo uses a chain of specialized AI agents (powered by GPT-5 reasoning models) to act as a coach:

NLP Logger: You don't toggle complex UI buttons. You just write a note: "Shoulder hurt on bench press, RPE was 9". The NLP agent parses this, logs the RPE, flags the injury, and automatically swaps the next push exercise to a joint-friendly variation (e.g., Landmine Press) for the next session. Methodology Compliance: The logic engine respects strict rules. If you select "Mentzer HIT", it enforces low volume/absolute failure. If you select "Kuba", it manages volume accumulation. [1] * Contextual Timer: It doesn’t just count down 90s. If the protocol is FST-7, it enforces short rest (30-45s) for fascial stretch. Current Status: We are in Open Beta.[2] The backend is heavily reliant on the new reasoning capabilities of GPT-5 to handle the multi-variable logic of periodization without hallucinating absurd weights.

I’d love you to roast the landing page or try the beta. Specifically interested in feedback on the NLP parsing accuracy from any serious lifters here.

Link: https://arvo.guru

Thanks!