Building Conversational AI with WebRTC and LLMs: Real-Time Voice Agents
A technical guide to building real-time voice AI agents using WebRTC for audio transport, speech-to-text, LLM reasoning, and text-to-speech in a low-latency pipeline.
6 min read
Agentic AI & LLMs
Agentic AI & LLMs archive page 537 of 631
Agentic AI & LLMs
Agentic AI, LLM engineering, and the models behind modern automation — multi-agent systems, LLM evaluation and comparisons, RAG, fine-tuning, AI infrastructure, security, and production AI engineering.
From the blog
Latest articles
Real-Time AI: Streaming, WebSockets, and Server-Sent Events for LLM Applications
How to build responsive AI applications using streaming, WebSockets, and SSE, with practical patterns for token streaming, agent status updates, and real-time collaboration.
5 min read
Where Agent Skills are heading next and how to prepare
Where Claude's agentic stack is heading — shared skill ecosystems, standard composition, governance — and the hygiene to prepare for it without over-betting.
7 min read
Measuring success for Claude Agent Skills in production
The real metrics for Claude Agent Skills — autonomy rate, rework, escalation quality, tokens per task, and eval pass rate — that prove agentic AI works.
7 min read
Skills in action: shipping a refund agent with Claude
An end-to-end walkthrough of building a Claude refund agent with Skills, MCP, subagents, and projects — from messy problem to safely shipped automation.
7 min read
Managing the risk of Claude Agent Skills in production
Failure scenarios, blast radius, and containment for Claude Agent Skills — scoping, approval gates, sandboxing, and staged rollout for safe agentic AI.
8 min read
Agent Skills: the new roles teams must hire and train
What engineers, integrators, and leaders must learn for Claude Agent Skills to work in production — and the new roles that emerge around them.
8 min read
Scaling Claude Skills from one team to the whole org
How to scale Claude Agent Skills from one team to the whole org without chaos — catalog, tiering, federated ownership, and failure modes to avoid.
6 min read
When to use Claude Skills vs prompts, MCP, or subagents
An honest decision guide for choosing Claude Agent Skills versus prompts, MCP servers, Projects, or subagents — and when not to build a skill.
7 min read