e911, Kari's Law, and RAY BAUM Compliance for AI Voice in 2026
What direct 911 dialing and dispatchable location requirements mean for AI receptionists running on multi-line telephone systems, and the configuration that satisfies the FCC.
9 min read
Agentic AI & LLMs
Agentic AI & LLMs archive page 290 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
Cost Monitoring for Token-Burn Outliers in Voice and Chat Agents
Mean token cost lies. Cost distributions are right-skewed and a single runaway agent can blow your monthly budget. Z-score and IQR alerts in 2026 catch the spike at minute 5, not month-end.
10 min read
SIP Debugging with sngrep and Wireshark for AI Voice Calls in 2026: The Hands-On Playbook
When your AI voice agent gets one-way audio, missed DTMF, or codec mismatch, sngrep and Wireshark are still the fastest path to root cause in 2026. Here is the playbook.
10 min read
Arize Phoenix in 2026: The Self-Hosted Observability Stack for Agents
Phoenix accepts OpenTelemetry traces from any framework, runs anywhere from your laptop to Kubernetes, and is fully open source with no feature gates.
11 min read
Migrate Twilio Studio/Flex Flows to an OpenAI Realtime Hybrid
Studio and Flex are great UI but bottleneck on rigid IVR logic. Add OpenAI Realtime as the natural-language frontend — keep Flex for human routing.
13 min read
Realtime Alerting on Call Sentiment Drops: A Pipeline That Actually Pages People in 2026
Sentiment alerting is easy to ship and hard to make useful. We cover thresholding, debouncing, baseline drift, and a Slack/PagerDuty integration that doesn't generate alert fatigue. Includes the SQL we use at CallSphere.
10 min read
Fireworks.ai for Voice Agents: FireAttention 4× Lower Latency (2026)
Fireworks.ai's proprietary FireAttention engine delivers 4× lower latency than vLLM, 150ms P50 TTFT on Llama 70B, and 92.1% multi-tool function calling accuracy. Voice-agent build guide.
12 min read
Postgres + DuckDB for AI Analytics: pg_duckdb Speeds Up OLAP 100x (2026)
pg_duckdb embeds DuckDB inside Postgres so transactional and analytic queries share the same database. AI dashboards that took 90 sec on Postgres run in <1 sec via DuckDB — without leaving Postgres.
11 min read
Pre-Fetching Common Tool Results for Voice Agents (2026)
Most voice-agent tool calls hit the same hot data: caller account, upcoming appointments, recent invoices. Pre-fetch on call connect so the LLM never waits. ToolCacheAgent and Asteria show 1.8-3.2x speedups.
11 min read