Otto

Meet Beam. Your product catalog is a mess. AI agents can't read it. Customers can't search it. Your data is stuck in 2010. Beam fixes all of it. ↳ Raw product data goes in ↳ AI-enriched, structured, searchable data comes out ↳ Every AI shopping agent can now find your products One API. Any catalog. Every discovery channel. Think of it this way: Stripe made payments easy for developers. Beam does the same for product data. We're building the developer infrastructure for agentic commerce. OpenAI launched ACP. Google launched UCP. Anthropic backs MCP. They built the rails. We're building the train. Waitlist is open → beamhq.io

#Beam 's entire MCP server is powered by #Laravel MCP SDK. It's surprisingly easy, clean and straightforward to expose MCP servers with Laravel. Five tools and any AI assistant can search, filter, and discover products from a Beam-powered catalog.✅ If anyones interested, here's docs: laravel.com/ai/mcp #laravel #beam #buildinpublic

#Beam WIP 189 tests. 600 assertions. I vibe code a lot of features but I always go back and read every test. Even if AI wrote it. (Especially if AI wrote it) That's honestly the fastest way to understand what your code actually does. #beam #buildinpublic #agenticcommerce

#Beam update! 🚀 Multi-project support is now live ✅ One account → multiple stores, brands, or client projects. Perfect for agencies, multi-brand merchants, or teams that actually want to collaborate. Team invitations drop next. We’re getting dangerously close. Watch demo and join the waitlist if you want early access: beamhq.io Who runs multiple stores or clients? 👇 #agenticcomerce #ecommerce #aicommerce #ucp #ACP #buildinpublic

@rezoundous I don't think it's that easy. We still go standard interview process where you check deep programming , system design and engineering understanding. If both passes, then pick who is better with AI tools or pick who doesn't use AI tools and teach to properly use AI

Which would you hire? 1. A developer who relies heavily on AI tools 2. A developer who doesn't use AI tools

@marclou @stripe You can checkout sumsub.com/pricing/ it costs 1.35$ per verification Or didit.me/pricing/ If you want to go cheaper

✅ 1,000+ verified buyers & sellers on TrustMRR It costs ~$800/mo with @stripe KYC API, though, so I'm looking for a cheaper alternative because it's my #1 most expensive tech stack now lol. Any recommendation?

@ProductHunt The developer infrastructure for agentic commerce. beamhq.io

gm legends! whatcha shipping this week? drop it in the replies 🫶

@dkshbuilds Yeah and it's very annoying already 😂 I'm not against vibe coding entire landing page (I do that as well) but cmon people, spend some hours polishing it to be somewhat different at least

99% of vibe coded websites look like this. Agreed ?

@eliana_jordan Building Beam. Takes raw product data and enriches it with AI — attributes, use cases, discovery tags, semantic search. API-first. beamhq.io

What are you building today? Please tell me it’s not another habit tracker app

@kabilanBuilds Working on #Beam. Think of it as the data intelligence layer between ecommerce stores and AI shopping agents. beamhq.io

What you're building today? Share here will give you a genuine feedback!

@aryanlabde Building Beam — making ecommerce product data actually readable for AI shopping agents. The kind of infrastructure nobody thinks about until AI agents can't find your products. beamhq.io

What are you guys working on this Sunday? Pitch your product. Get some eyeballs to it.

@jonathan_wilke Building #Beam — making ecommerce product data actually readable for AI shopping agents. The kind of infrastructure nobody thinks about until AI agents can't find your products. beamhq.io

Raise your hand if you're building on Sundays ✋ Drop your project link below - let's see what you're cooking.

Hey Everyone ! Been building #Beam — an API-first platform that takes messy ecommerce product data and makes it AI-agent-ready. Enrichment, semantic search, recommendations, all through one API. The problem: AI shopping agents (OpenAI's ACP, Google's UCP, Anthropic's MCP) all need structured product data. Most stores don't have it. Just got the landing page live: beamhq.io Building solo, still early. Would love honest feedback — does it make sense? Anything confusing? Roast it if you want.

#Beam now has a home. Finally got the landing page and waitlist live. Kept it clean — no overused AI gradients, just a clear explanation of what Beam does and why it matters. Product demo coming soon 🚀🚀🚀 beamhq.io

Little sneak peak of Beam AI Platform, we are getting close but still lot of stuff to do and tweak 🚀🚀🚀🚀 #aitools #agenticcommerce #ecommerceai #beamhq

One of the hardest things about building a developer tool isn't the product. It's the landing page. How do you show enrichment, semantic search, recommendations, chat, and multi-protocol distribution — without it looking like a cluttered mess? And no, AI-generated landing pages don't help. They all look the same. Same gradients. Same "Powered by AI" hero. Same generic SaaS template energy.

"something comfortable for marathon training on wet roads, under $150, not Nike" Try putting that into any ecommerce search bar. Now imagine an AI agent trying to find that product for you. Traditional search matches keywords. AI agents need to understand attributes, context, intent, and constraints — all at once. Most product catalogs aren't built for that. That's the infrastructure gap nobody's talking about.

So I’m building it. An API-first platform that takes raw product catalogs and turns them into AI-agent-ready data — enriched, structured, and searchable across ACP, UCP, and MCP. Building in public. More soon. If you’re thinking about agentic commerce, this is going to matter.

I looked at who’s solving this. Enterprise platforms → too expensive, too slow Shopify apps → locked to one ecosystem Chat widgets → surface-level, no infra But if you’re a developer who wants to build agentic commerce into YOUR product? There’s nothing. Stripe exists for payments. Nothing like it exists for product data.

Most ecommerce product data is unreadable for AI. I noticed this months ago — if you try to build RAG-based shopping agents, retrieval falls apart because the underlying data is garbage. Missing attributes, inconsistent formats, zero structure. A month later Google launched UCP and confirmed exactly this. #agenticcommerce #aicommerce #rag