Runner AI
Build, optimize, and scale your AI-native store
559 followers
Build, optimize, and scale your AI-native store
559 followers
Don't just prompt a website; prompt revenue. Runner AI builds AND optimizes. It continuously runs experiments in the background, automatically turning visitors into buyers.
Free Options
Launch Team
CloudflareYour next app belongs on Cloudflare Workers. Deploy for free
Promoted
Hey Product Hunt!
I'm Weizhi, founder of Runner AI. Excited to share what we've been building over the past 5 months: an AI-native commerce stack that autonomously builds, tests, and optimizes your store.
Before starting Runner AI, I spent five years at Google and DeepMind — building data infra and Gemini. That shaped how I think about products: when data is unified and AI can see the full picture, it doesn't just help — it proposes changes, learns from real user behavior, and gets smarter with every iteration.
But today, traditional e-commerce doesn't work that way. They stitch together 20+ tools that don't talk to each other. Your email tool doesn't know what your checkout learned. Your analytics can't inform your pricing. Because nothing connects, nothing compounds. You're running experiments in silos while customer behavior shifts daily.
We built Runner to bring that self-optimizing logic to commerce, grounded in the belief that the best way to grow is to experiment, learn, and iterate continuously.
Most e-commerce platforms are built for humans to operate, with AI added later as a feature. We designed it differently — a commerce stack built from the ground up for AI to observe, act, and improve your conversion rates automatically.
Runner puts everything under one roof: storefront, analytics, experiments, pricing, promotions, user behavior — all unified, all visible to AI. When checkout improves, recommendations get smarter. When pricing learns, promotions adapt.
How it works:
Describe & Build: Describe your vision → AI builds everything (design, copy, functionality).
Observe: AI sees everything: drop-offs, abandoned products, winning copy, and pricing friction.
Optimize: AI acts on everything: rewrites headlines, adjusts layouts, optimizes flows, and tests pricing.
Scale: Winners get scaled. Losers get replaced. 24/7.
What makes us different:
No themes, no templates: AI architects a custom store from scratch.
No "App Tax": Core tools are built-in, not paid plugins.
Full transparency & control: You see every experiment, set boundaries, and adjust anytime.
This is v1. We'd love for you to try it and tell us what you think.
→ Try Runner AI free
Weizhi Li
Founder & CEO, Runner AI
@weizhi So incredibly proud to be part of this team and what we've built over the last few months! Watching this vision of autonomous commerce come to life has been amazing. Let's go!
@stacy_liu2 let's go!!!
@weizhi What a journey getting here! As an engineering team, our goal was to build a truly AI-native architecture from day one. We didn't want to just bolt an LLM onto a legacy system; we wanted a self-optimizing engine capable of continuous, agentic commerce. The infra is built to scale, learn, and iterate 24/7 without human intervention. So proud of what the team has shipped today!
@runner_yufeng Couldn't have said it better! The engineering work behind this has been incredible. Proud of the team for making it happen! Excited for what we ship next!
The "continuous experiments in the background" angle is really compelling — most A/B testing tools require you to manually set up experiments which nobody actually does consistently.
Curious how it handles low traffic sites where statistical significance takes forever to reach? That's always been the frustration with traditional CRO tools for early-stage products.
Congrats on the launch!
@n8nship You are right. low traffic is where traditional A/B testing breaks down. Our AI uses multi-armed bandit approaches that adapt in real time, so even with lower traffic, the system learns and shifts toward better-performing variants faster instead of waiting for a fixed sample size. On top of that, we also have built-in marketing integrations — ads, social media, and more — to help drive traffic to your store in the first place.
The "self-optimizing" framing is what caught my attention here. Most ecommerce AI tools I've seen are basically just suggestions you still have to manually implement. Having the system actually run experiments and adjust on its own is a totally different level.
One question though, how much control do store owners have over what gets tested? Like can you set guardrails so it doesn't experiment with things you'd rather keep fixed (brand elements, pricing floors, etc)?
@mihir_kanzariya Great question! You define the boundaries — pricing, visuals, ads, social media, promotions, etc. You also control the traffic split, so you decide how much of your audience sees a test before it rolls out.
@mihir_kanzariya You are absolutely still in the driver's seat. While Runner AI analyzes every interaction on your site (clicks, scrolls, bounces) to detect friction and automatically plan A/B tests, you have full control over the boundaries. You can easily set strict guardrails to lock in brand elements, copy, or pricing floors so the AI knows exactly what not to touch. You can also dictate exactly how much of your traffic you want to allocate to these experiments.
Awesome!
Are you helping stores with product feed optimization for agentic commerce?
@kaiserrr Absolutely! Runner AI continuously detects friction and autonomously A/B tests to optimize the product feed. Because our entire architecture is AI-native from the ground up, your product data is perfectly structured for agentic commerce and AI shopping bots by default.
@stacy_liu2 Awesome!
We should chat after the launch. Atb!
@stacy_liu2 @kaiserrr yes. let's chat
@kaiserrr Absolutely! Product feed optimization is built right in. We handle formatting across OpenAI ACP and Google UCP — so your store is ready for agentic commerce out of the box.
The self-optimizing angle is really interesting — most e-commerce tools are static but this actually adapts. Curious how it handles product catalog updates automatically? Congrats on the launch! 🚀
@dheeraj_nt Thanks! Great question. there's a full transactional system backing the platform, so when you update your catalog, everything syncs automatically: storefront, product feeds, recommendations, and any running experiments.
Congrats on the launch!
If Runner AI is running experiments automatically, how do you handle guardrails for sensitive things like pricing or checkout changes?
@hmadhsan Yes, all the non visual changes made are backed by our transaction system. and guardrails are applied there
Congratulations on the launch! Can I control exactly what your AI changes? For example, if I ordered SEO page optimization, made edits to the pages, and after that I don’t need it anymore. Your AI won’t accidentally change it without my knowledge, right?
@mykyta_semenov_ That's great point. yes, you're always in control. You set the boundaries and the AI only operates within them. i.e. pages in your example