I mean a system where one can do 10k calls per day. Has anyone built a system like this using livekit and pipecat. did you do it without using our own GPUs ?

Do we have any opensource orchestration platforms built on Livekit that work similar to vapi or Retell?

While wrappers are great for MVPs, building your own orchestration layer gives you 𝗳𝘂𝗹𝗹 𝗼𝘄𝗻𝗲𝗿𝘀𝗵𝗶𝗽, 𝘀𝗶𝗴𝗻𝗶𝗳𝗶𝗰𝗮𝗻𝘁𝗹𝘆 𝗹𝗼𝘄𝗲𝗿 𝗰𝗼𝘀𝘁𝘀, 𝗮𝗻𝗱 𝗴𝗿𝗮𝗻𝘂𝗹𝗮𝗿 𝗰𝗼𝗻𝘁𝗿𝗼𝗹 over the entire conversational pipeline. I designed this engine to fully replace third-party wrappers like Vapi & Retell AI. Here is a deep dive into what’s under the hood: 🔄 𝗗𝘆𝗻𝗮𝗺𝗶𝗰 𝗔𝗴𝗲𝗻𝘁 𝗖𝗼𝗻𝗳𝗶𝗴𝘂𝗿𝗮𝘁𝗶𝗼𝗻 (𝗥𝗲𝗮𝗹-𝗧𝗶𝗺𝗲 𝗛𝘆𝗱𝗿𝗮𝘁𝗶𝗼𝗻) Hardcoding agents is a trap. I implemented a system that executes an API call upon call initialization. • 𝗛𝗼𝘁-𝗦𝘄𝗮𝗽𝗽𝗮𝗯𝗹𝗲 𝗣𝗲𝗿𝘀𝗼𝗻𝗮𝘀: A single engine instance can instantly apply unique System Prompts, Voice IDs, and Temperature settings based on backend parameters. • 𝗥𝗲𝘀𝘂𝗹𝘁: You can power thousands of unique agents (e.g., specific to different businesses) without ever redeploying the core code or creating a new instance. 🛠️ 𝗖𝗼𝗻𝘁𝗲𝘅𝘁-𝗔𝘄𝗮𝗿𝗲 𝗙𝘂𝗻𝗰𝘁𝗶𝗼𝗻 𝗥𝗼𝘂𝘁𝗲𝗿 When building raw infrastructure, manually mapping tools to agents is a major architectural hassle. I built specialized helper logic for 𝗗𝘆𝗻𝗮𝗺𝗶𝗰 𝗧𝗼𝗼𝗹 𝗜𝗻𝗷𝗲𝗰𝘁𝗶𝗼𝗻 to solve this. • 𝗠𝗼𝗱𝘂𝗹𝗮𝗿 𝗟𝗼𝗴𝗶𝗰: The router decouples the orchestration engine from business logic. It parses the backend setup and assigns only the specific tools defined in that agent's configuration (e.g., loading "Appointment Booking" tools only when the specific use-case demands it). 💾 𝗗𝗮𝘁𝗮 𝗣𝗲𝗿𝘀𝗶𝘀𝘁𝗲𝗻𝗰𝗲 & 𝗣𝗼𝘀𝘁-𝗖𝗮𝗹𝗹 𝗜𝗻𝘁𝗲𝗹𝗹𝗶𝗴𝗲𝗻𝗰𝗲 Logs aren't enough. I built a save_conversation function that aggregates the full session payload and triggers intelligent sub-functions immediately after the call: • 𝗖𝗮𝗹𝗹 𝗦𝘂𝗺𝗺𝗮𝗿𝘆: Generates a natural language recap via LLM. • 𝗖𝗮𝗹𝗹 𝗘𝘃𝗮𝗹𝘂𝗮𝘁𝗶𝗼𝗻: Structurally classifies the outcome (e.g., "Booked", "Inquiry", "Failed"). • 𝗧𝗲𝗹𝗲𝗺𝗲𝘁𝗿𝘆: Captures precise Token Usage (for billing) and Latency statistics alongside the transcript. 🛡️ 𝗣𝗿𝗼𝗱𝘂𝗰𝘁𝗶𝗼𝗻 𝗚𝘂𝗮𝗿𝗱𝗿𝗮𝗶𝗹𝘀 To prevent runaway costs and "zombie" connections, I engineered active background monitors: • 𝗜𝗻𝗮𝗰𝘁𝗶𝘃𝗶𝘁𝘆 𝗠𝗼𝗻𝗶𝘁𝗼𝗿: Detects silence (30s default) and gracefully terminates the session.

I’m planning a few LiveKit and Pipecat live sessions over the next weeks, and I’d love to open them up for community contributions. If you’re interested in hosting a session or sharing your expertise, feel free to DM me. Here are some topic ideas to spark inspiration: - Latency optimization: Strategies to achieve sub-600 ms latency - Interruption handling - Industry-specific use cases: Real estate, dental, medical, HVAC, restaurants, hotels, etc. - Integrations with niche software that rarely gets covered: ServiceTitan, FieldEdge, and Housecall Pro for HVAC and home services; Dentrix, Open Dental, and Eaglesoft for dental practices ; Epic, Athenahealth, and Oracle Health for medical practices; Toast, Aloha, and Square for restaurants; Opera PMS, Cloudbeds, and Mews for hotels; Buildium, Propertyware, and Follow Up Boss for real estate and property management. - Telephony provider integrations beyond Twilio: Telnyx, Zadarma, RingCentral, Ringba - Custom PBX integrations - Multilingual implementations: Best voices for specific languages, best transcription models for specific languages and best prompting strategies, deep dives into a single language - Non-technical Voice AI topics: Project management, hiring and evaluating developers, marketing Voice AI products and services, finding clients, content creation, proposal & contract creation, compliance You’re also welcome to use your session to showcase your own product or service, as long as it aligns with the theme of open-source Voice AI.

You’re probably still building AI voice agents manually for clients. That’s not a scalable business. It’s time to take your AI agency to the next level by launching your own full-stack SaaS platform — just like I did. The good news: I’ve already built the entire AI voice agent SaaS platform, so you don’t have to. Here’s what’s inside the codebase: 🗣 AI Voice Agent — powered by Vapi & LiveKit⚙ Configurations — prompts, models, knowledge base, and more 📝 Call Logs 📈 Analytics Dashboard 🔐 Sign in / Sign up — Google Authentication ready 💳 Payment Collection — via Stripe 🏢 Multi-Tenant Architecture — one user can be linked to multiple organizations 📞 SIP Connection — integrated with Telnyx ⛔ Rate Limiter — manage usage efficiently And everything else you need to launch a production-ready SaaS platform. 💬 Comment below or DM me if you’re interested in using or acquiring the full codebase for your own AI voice agent business.