befday

The company described as "befday" appears to be a self-reported project built for OpenAI Build Week 2026. The author states that it helps people discover birthday perks from local shops and gives those shops AI agents, inbox replies, vouchers, and loyalty tools. It is presented as an extension of an existing platform with new AI-powered features for merchants.

The core commercial read: the project is a demo built around a single merchant (Sunrise Café) to showcase how AI can be used in local business operations — particularly for customer engagement via birthday perks and automated inbox replies.

The description states that befday helps customers discover birthday perks from local shops, and gives merchants:

• A dashboard for vouchers, loyalty, POS, and a team inbox
• AI agents configured with custom instructions, reply modes (draft / hybrid), and knowledge sources (FAQs, docs, website)
• Draft or hybrid inbox replies grounded in each shop’s own knowledge — approve, edit, or discard before send
• Inbound workflows with branching, HTTP steps, wait_for_reply, language preference, and inbox-triggered published flows

The author describes the product as built using OpenAI Codex (5.3 medium) and GPT-5.6 during Build Week, with a focus on:

• AI agent builder
• Knowledge ingestion and grounding
• Inbox hybrid replies
• Inbound workflow runtime

Not evidenced: No mention of actual users, customers, or revenue.

The author states that befday began as a way to help people discover birthday perks from nearby merchants, but evolved into a platform that gives local shops AI agents and automation tools.

It positions itself as:

• A tool for personalizing customer relationships like big brands
• An extension of an existing product (previously had discovery, vouchers, POS, loyalty)
• Focused on reliability over novelty, with hybrid/draft replies to maintain trust

Not evidenced: No claims about market traction, competitive positioning, or prior user feedback.

The description states that local shops are the primary target, especially those who know their regulars but lack tools for personalization.

It also mentions:

• Merchants can configure AI agents with custom instructions and knowledge sources
• The system supports inbox replies, workflows, and loyalty features

Not evidenced: No information on actual customer segments, size of target market, or specific merchant personas.

The description does not state anything about pricing, monetization, or business model.

It mentions that merchants get access to:

• Dashboard for vouchers, loyalty, POS, and a team inbox
• AI agents with configuration options
• Inbox replies and workflows

Not evidenced: No pricing structure, revenue streams, or commercial terms.

The author states that the project was built using:

• OpenAI Codex (5.3 medium) and GPT-5.6
• Tools: aisdk, better-auth, bun, codex, cursor, drizzle, eve, expo.io, firecrawl, next.js, openai, postgresql, react, tailwind-css, trpc, turborepo, typescript, vercel, workflow

The system supports:

• AI agent builder with publish/suggested actions
• Knowledge ingestion and grounding (FAQ/docs upload, embeddings, Firecrawl Map page selection)
• Inbox hybrid replies with approval drafts vs send
• Inbound workflows with branching, HTTP steps, wait_for_reply, delays, nested runs

Not evidenced: No information on scalability, infrastructure, or deployment details beyond the tech stack.

The description states that:

• The project was built for OpenAI Build Week 2026
• It includes a live demo path: customer asks “Whats the wifi password?” on Telegram → Inbound → AI Agent runs → Sunrise Café Assistant drafts the answer from FAQ → merchant approves before send
• A demo shop (Sunrise Café) is used to run end-to-end in minutes

Not evidenced: No data on actual users, adoption rates, or product usage beyond a demo.

The description does not mention any competitors or competitive landscape.

It implies that local shops lack tools for personalization, and that the solution aims to bridge this gap using AI agents and automation.

Not evidenced: No information on existing solutions in the marketplace, market size, or competitive differentiation.

• The project is described as a demo built during a hackathon — no evidence of real-world usage or traction
• The author states that the platform already existed before Build Week and that new features were added; this raises questions about whether the product is truly ready for market or still in development
• There’s no mention of commercial viability, pricing, or monetization strategy
• The focus on hybrid/draft replies suggests a cautious approach to AI automation — but also implies that full autonomy may not be feasible or desired by merchants

Not evidenced: No data on risks related to product-market fit, scalability, or long-term sustainability.

1. What was the pre-existing platform like before Build Week? Was it used by real merchants?
2. How many merchants are currently using the platform beyond the demo?
3. Is there a plan for monetization or pricing structure?
4. What is the roadmap for moving from demo to production-ready product?
5. How does the hybrid reply mode affect customer experience and merchant adoption?
6. Are there any plans to expand beyond birthday perks or local shops?