I built an AI front desk for the Meow Meow Veterinary Hospital competition round. You type what you need in your own words, and it books, cancels or reschedules against a live Google Calendar. Every request is logged to Google Sheets, and it sends a confirmation email. The design rule: the LLM only turns messy text into structured JSON. All the rules and writes are deterministic code. A misread message can cause a question or a refusal, but never a bad booking. The problem I hit: my form test passed, but the public form threw “Workflow could not be started” in a real browser. n8n won’t start a form workflow that also has a Respond to Webhook node. My test went through the API, which skips that check, so it was a false pass. How I solved it: 1. Split it into a form workflow and a main pipeline workflow. 2. The form workflow passes fields to the main one and shows the HTML result. 3. Re-verified it end to end, and wrote the correction into my test report instead of hiding it. Other bugs live testing caught: - The email step crashed after the booking was saved. It used a helper the n8n code sandbox doesn’t have. I made both email steps crash-proof. - “Wellness exam for a bird” booked an Exotic Exam. A prompt fix plus a rule check solved it. What I’m proud of: - 20 of 20 required tests and 14 extra cases passing live - Rollback if the sheet write fails after the calendar event is created - Replay-safe: the same request ID returns the original answer instead of double-booking - Prompt injection and medical-advice requests are refused or handed off Lesson: a test only counts if it uses the same path your users do. Screenshots are attached: the workflow canvas, the form, and the records sheet. Happy to share the structure or prompts if anyone wants them. How do you test your n8n workflows? 👇

What if multiple AI agents could work together to turn a game concept into a playable prototype? We're exploring a multi-agent AI system where each agent handles a specific part of the game development pipeline, from planning and 3D asset creation to level assembly, gameplay implementation, and testing. 🤖 How the system will work Our proposed agent team includes: - Project Manager Agent: Understands the brief, defines scope, and coordinates tasks. - Game Designer Agent: Plans the level, mission objectives, gameplay rules, and spawn points. - 3D Asset Agent (Blender): Creates or prepares 3D models, materials, and textures. - Environment & Level Agent (Unreal Engine): Assembles modular assets and builds the game environment. - Gameplay Logic Agent: Implements gameplay systems using Blueprints and C++ where needed. - QA & Testing Agent: Runs tests, identifies issues, and validates gameplay. - Documentation Agent: Tracks changes, decisions, errors, and development progress. ⚙️ Proposed technology stack - Claude Code + Claude Opus: Reasoning, planning, coding, and agent coordination. - Blender MCP + Unreal MCP: Connecting AI workflows to the respective development tools, subject to integration testing. - Unreal Engine 5.8: Level assembly, gameplay, physics, and rendering. - Python: Task orchestration, structured specifications, and automation. - Git + Git LFS: Version control and recoverable checkpoints. - Unreal testing framework: Gameplay validation and regression checks. 🔄 Development workflow Game Brief → Planning → Asset Creation in Blender → Import into Unreal → Level Assembly → Gameplay Integration → Testing & Validation → Playable Build Each stage will produce trackable outputs, with logs, checkpoints, and bounded repair attempts to help make the process reproducible and recoverable. 🏗️ Our initial goal We're starting with a small, playable military-style prototype featuring an industrial outpost, two teams, a capture-point objective, one weapon, health, reload, respawn, and up to four bots.

How I Built an AI Voice Receptionist That Actually Books Appointments (Under 800ms Latency) Hey everyone! 👋 70%+ of after-hours callers hang up the second they hear a voicemail beep—and book with the next competitor on Google. Alongside our platform work at Oqvera, I’m expanding my custom voice AI and automation services to Fiverr. Here is the exact production blueprint I use to automate inbound calls and eliminate missed leads: 🛠️ The Core Stack - Voice & Telephony: Vapi AI / Retell AI + Twilio - STT & Audio: Deepgram Nova-2 + Cartesia / ElevenLabs - LLM Engine: GPT-4o / Claude 3.5 Sonnet (Tool Calling) - Automation & CRM: n8n / Make.com + GoHighLevel / Cal.com ⚙️ How the System Operates 1. Sub-800ms Inbound Response: Instant, human-like voice handling with zero awkward robotic pauses. 2. Dynamic Live Scheduling: AI queries live calendar slots via API in real time, avoids double bookings, and reserves appointments on the spot. 3. Guardrailed FAQs: Answers pricing, location, and service questions with zero hallucinations. 4. Automated Post-Call Sync: n8n instantly pushes call transcripts to CRM, fires an SMS confirmation to the caller, and initiates warm transfers for emergency cases. 💡 Pro Tip Keep total round-trip latency under 1 second and always set up an SMS/transfer fallback for edge cases. If you want a ready-to-deploy voice receptionist for your business or clients: 👉 Check out my Fiverr Gig: https://www.fiverr.com/s/AGBz4qX For the voice builders here: Vapi or Retell AI—what is your go-to framework right now? Let’s talk below!

Excited to share an update on a project I've been building — an AI-powered receptionist for auto dealership showrooms. The problem it solves: dealerships lose leads constantly because no one answers fast enough — after-hours calls go to voicemail, web leads sit overnight, and by the time someone follows up, the customer's already talking to a competitor. What it does: → Answers calls & chat instantly, 24/7 → Qualifies leads (vehicle interest, trade-in, timeline) → Books test drives/appointments straight into the calendar → Checks live inventory availability → Auto-follows up on cold leads so nothing slips through → Hands off warm leads to a real rep with full context I've built it out for a client and it's now live in the testing phase — running real calls, measuring response times and booked appointments before we roll it out further. More updates soon as the pilot data comes in. Happy to answer questions if anyone's exploring something similar in their own niche 👇

we are looking for ai expert