Dominic Holt

Shipping today: Cartesia’s Ink-Whisper on Vapi. Latency is critical. Slow STT ruins conversations. With Ink-Whisper, just one line in your Vapi config gets you: - Faster, streaming-optimized transcripts - Better handling of background noise and accents Read more: https://vapi.sh/u4KRN67

111

11 Comments

When’s the last time your LLM provider went down? 🤔 Well, OpenAI had a partial outage late Monday night and Anthropic had elevated error rates as recently as yesterday. At Assembled, our customers don’t tolerate downtime like this, so we built automated fallbacks, making our system incredibly resilient to LLM provider outages without our customers needing to know. We detect failures in real time, instantly reroute requests to secondary or on-prem models, and keep your support workflows humming—even when the big providers falter. 🚀 We wrote a blog post on the nuts and bolts of how we: - Monitor health across multiple LLM providers - Seamlessly fail over to backup models or local inference - Maintain sub-second response times and 99.99% uptime Check it out here 👉 https://lnkd.in/g3s7y4fc

58

4 Comments

Just rewatched Paz's presentation at CMX — one of the more thoughtful breakdowns I’ve seen on scaling community. She shared how she used Intros AI 🤝 to automate different types of member introductions, making it easy to create meaningful connections without manual coordination. One line stuck with me: “We connected hundreds of people across 12 countries in minutes.” As community building evolves, so should our tools. Automating logistics isn’t about removing the human — it’s about making space for deeper, more intentional engagement.

37

4 Comments

🚀 OpenAI 𝐣𝐮𝐬𝐭 𝐜𝐡𝐚𝐧𝐠𝐞𝐝 𝐞𝐯𝐞𝐫𝐲𝐭𝐡𝐢𝐧𝐠 (𝐚𝐠𝐚𝐢𝐧) with the launch of 𝐀𝐠𝐞𝐧𝐭𝐊𝐢𝐭 on Dev Day 2025 - and the implications for the agent ecosystem are massive. For years, building true “agents” has required stitching together frameworks like LangChain, CrewAI or 𝐀𝐃𝐊𝐬 - each great in its own right, but all requiring manual plumbing, action wiring, and environment setup. AgentKit flips that model. It brings the first 𝐧𝐚𝐭𝐢𝐯𝐞 𝐚𝐠𝐞𝐧𝐭 𝐫𝐮𝐧𝐭𝐢𝐦𝐞 directly inside OpenAI’s ecosystem - one where GPTs can 𝐞𝐱𝐞𝐜𝐮𝐭𝐞 𝐭𝐨𝐨𝐥𝐬, 𝐜𝐚𝐥𝐥 𝐀𝐏𝐈𝐬, 𝐚𝐧𝐝 𝐩𝐞𝐫𝐬𝐢𝐬𝐭 𝐬𝐭𝐚𝐭𝐞 𝐬𝐞𝐚𝐦𝐥𝐞𝐬𝐬𝐥𝐲. This is OpenAI’s most serious move yet toward agents that can actually do things, not just talk about them. But here’s the twist: 👉 Frameworks/ Toolkits like 𝐋𝐚𝐧𝐠𝐂𝐡𝐚𝐢𝐧, 𝐀𝐠𝐞𝐧𝐭𝐊𝐢𝐭 and 𝐀𝐃𝐊𝐬 still depend on developers to manually define the “what” (tools, APIs, and actions). 👉 Adopt AI 𝐚𝐮𝐭𝐨𝐦𝐚𝐭𝐞𝐬 𝐭𝐡𝐚𝐭 𝐥𝐚𝐲𝐞𝐫 𝐜𝐨𝐦𝐩𝐥𝐞𝐭𝐞𝐥𝐲. With 𝐀𝐝𝐨𝐩𝐭 + 𝐀𝐠𝐞𝐧𝐭𝐊𝐢𝐭, you get the best of both worlds: ⚙️ 𝐀𝐠𝐞𝐧𝐭𝐊𝐢𝐭 gives your LLMs execution and persistence. 🧠 𝐀𝐝𝐨𝐩𝐭 auto-discovers APIs, generates actions, and makes any application—legacy or modern—agent-ready instantly. Together, they collapse the 𝟎 → 𝟏 𝐣𝐨𝐮𝐫𝐧𝐞𝐲 of building agents that can execute actions across any application into hours instead of months. This isn’t just another framework integration. It’s a step toward 𝐬𝐞𝐥𝐟-𝐜𝐨𝐧𝐟𝐢𝐠𝐮𝐫𝐢𝐧𝐠, 𝐬𝐞𝐥𝐟-𝐮𝐩𝐝𝐚𝐭𝐢𝐧𝐠 𝐚𝐠𝐞𝐧𝐭𝐬 that can reason, act, and evolve—without manual specs or scaffolding. ✨ 𝐓𝐡𝐞 𝐀𝐠𝐞𝐧𝐭𝐢𝐜 𝐞𝐫𝐚 𝐣𝐮𝐬𝐭 𝐠𝐨𝐭 𝐚 𝐦𝐚𝐣𝐨𝐫 𝐮𝐩𝐠𝐫𝐚𝐝𝐞. #AdoptAI #AgentKit #OpenAI #DevDay #AIagents

49

1 Comment