Guides
Start with the concepts that survive provider changes.
Learn the architecture, product categories, and evaluation vocabulary behind production conversational AI systems.
Open-source roadmap
Read the full field guide
A multilingual, vendor-neutral roadmap covering chat, voice, multimodal agents, telephony, evaluation, safety, and production.
Static benchmark v1
Compare the ecosystem
Browse platforms, realtime APIs, STT, TTS, and evaluation tools before you choose a stack.
Foundational articles
-
Conversational AI Evaluation Metrics
The metrics that matter for evaluating conversational AI systems, including task success, latency, tool reliability, safety, cost, and voice-specific timing.
-
How To Choose A Voice Agent Stack
A developer-first framework for choosing between managed voice platforms, open-source frameworks, realtime model APIs, and RTC or telephony infrastructure.
-
Voice Agents vs Chat Agents vs Multimodal Agents
How to choose between chat, voice, phone, and multimodal conversational AI interfaces based on user context, latency, risk, and workflow complexity.
-
Conversational AI Architecture
A practical architecture map for conversational AI systems, from text chat and RAG to realtime voice, telephony, multimodal sessions, tools, and evaluation.
-
What Is Conversational AI?
A practical developer definition of conversational AI across chat, voice, realtime, telephony, multimodal agents, and workflow automation.