LangGraph for Healthcare Prior-Auth Workflows: Production Story
How a Massachusetts payer is using LangGraph 1.0 to automate prior-authorization workflows with HITL, audit logs, and HIPAA-safe state for real volume.
10 min read
Agentic AI & LLMs
Agentic AI & LLMs archive page 260 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
Reranking Cuts RAG Cost 3x — With Trade-Offs You Should Know
Adding a reranker lets you retrieve more candidates and use a cheaper LLM downstream. The math, the latency cost, and where this approach quietly backfires in production.
10 min read
HIPAA and AI Training Data Exclusion: The Zero-Retention BAA in 2026
Most HIPAA-eligible AI deployments depend on one operational pattern: zero-data-retention endpoints with explicit training-data exclusion. Here is the contract language, the technical pattern, and the gotchas.
10 min read
Idempotency Keys for AI Tool Calls: Stripe-Style Safety When the Agent Retries
Network retries, queue redeliveries, and saga compensations all create double-execution risk. Idempotency keys at the tool boundary prevent your AI agent from booking the same slot twice.
10 min read
Build a Multi-Region Voice Agent on Fly.io for Sub-500ms Global Latency (2026)
Deploy a voice agent to Fly.io's anycast network across 6 regions: Tokyo, Frankfurt, São Paulo, Sydney, Virginia, Los Angeles. fly-replay routes traffic to the closest healthy region.
11 min read
WebRTC + AI Debate Moderator for Live Streaming in 2026: Real-Time Fact Checks
Live debate streams in 2026 ship an AI moderator that runs the timer, flags interruptions, and surfaces fact-checks via on-screen citations. Here is the WebRTC + Factiverse-style production stack.
10 min
Twilio Notify EOL: AI Multi-Channel Reach Without Notify (2026)
Twilio Notify reached end-of-life Dec 31 2025. We map the 2026 replacement stack — Conversations + Conversation Orchestrator + Messaging Service + push providers — and how CallSphere fans alerts to voice, SMS, WhatsApp, and email.
10 min read
Where production MCP agents are heading next
Agents reaching production via MCP are early. Where the capability is heading in 2026 and beyond, and how to prepare your team and architecture now.
7 min read
Measuring success for production MCP agents
Prove a production MCP agent is working: the metrics, eval discipline, and real-time signals that separate real success from a convincing demo.
7 min read