Phone Call Alerts for Prometheus Alertmanager and Grafana
Phone call alerts from Prometheus Alertmanager and Grafana: webhook receiver config, Grafana contact point, send_resolved, dedup and press-1 ack.
43 articles about technical. Discover expert insights, tutorials, and guides.
Phone call alerts from Prometheus Alertmanager and Grafana: webhook receiver config, Grafana contact point, send_resolved, dedup and press-1 ack.
Monitor cron jobs with a heartbeat URL: add a curl ping to crontab, report failures with ?status=fail, and get a phone call when a job doesn't run.
What an on-call escalation policy is: steps, wait times, call and email channels, repeats and acknowledgement, with example policies for small teams.
Add phone call notifications to self-hosted Uptime Kuma: a Webhook notification to Edesy OnCall, press-1 acknowledgement, Hindi or English calls and escalation.
Why uptime alerts should ring your phone: missed email and chat alerts, press-1 acknowledgement, escalation to a backup, and a free plan with 3 monitors.
What a voice API status callback is, which statuses a call goes through, how to read SIP codes, hangup reasons and call outcomes, and how to verify, deduplicate and replay callbacks.
Give a LiveKit Cloud voice agent an Indian phone number: create inbound and outbound SIP trunks with the lk CLI, add a dispatch rule, route the Edesy number, and test both directions.
How Conversation Relay works: your bot exchanges text over a WebSocket while Edesy handles speech recognition, text-to-speech, barge-in and keypresses on an Indian phone number. Twilio ConversationRelay-compatible.
Why SIP calls have one-way audio or drop at 30 seconds, and how to fix it: disable SIP ALG, set NAT external address, open the RTP range, and avoid codec mismatches.
How a voice OTP call works step by step: the API request, code generation, the spoken template, press-1 confirmation, retries, rate limits, the callback and what each call costs.
How to stream live phone call audio to your own voice bot over a WebSocket: the connection lifecycle, raw L16 PCM frames, barge-in, DTMF, sample rates, and how it compares with base64 mu-law streams.
How to connect an xAI Grok Voice Agent to an Indian phone number over SIP: register the number with xAI, save the Grok SIP URI on Edesy, route the number, and test without dialling.
Add automated COD confirmation calls to a custom store: send a signed order webhook, verify X-Edesy-Signature HMAC-SHA256, and handle confirm or cancel on your action URL. Payloads and code included.
Why AI voice calls go silent after "hello" or mid-call, from real incidents: missing model settings, silent tool calls, recording announcements.
Stop AI voice agents talking to call-screening assistants, "call is recorded" notices and network announcements. Prompt rules from real calls.
When to build an AI voice agent as a free-form prompt and when to use a node-based AI Flow, with order status, booking and qualification examples.
Lessons from a deployment where AI-to-receptionist transfers silently failed on one phone route. How to test, log and recover call transfers.
Why AI voice agents read placeholders aloud, keep only first names or skip names on Indian calls, and the prompt fix that captures full names reliably.
Ten prompt rules we add to every AI voice agent after real Indian deployments: closing, silence, names, numbers, one question per turn and more.
Real-deployment fixes for how AI voice agents say Indian numbers (lakh, crore), class names, dates, amounts and industry acronyms, so calls sound natural.
How to give an AI voice agent custom API tools so it can look up orders, check slots, create tickets or update records in your own system during a live call.
On Hindi-English calls, speech-to-text can write English acronyms in Devanagari and break keyword-based qualification. Why, and how to design around it.
When an AI voice agent should ask callers to press keys instead of speaking: confirmations, order numbers and noisy lines. Never for PINs or OTPs.
Build an AI voice agent that greets in two languages, detects the caller's language and hands off to a specialist for Hindi, Tamil, Telugu and more.
How real deployments classify every AI call into one-word outcomes like APPROVED, CALLBACK or WRONG_PERSON, and turn them into dashboard columns and CRM fields.
How to split one AI voice call across specialist agents (greeter, sales, support, billing) that hand off mid-call, each with focused instructions.
Native speech-to-speech vs the speech recognition, LLM and TTS pipeline for AI calls in India: naturalness, cost, reliability and lessons from production.
How to trigger AI payment reminder calls from your own app or billing system with one API call: pass the amount and due date, get the outcome back on a webhook.
How to verify a caller's claim on AI calls: open questions, never supplying the answer options, and scoring refusals. Lessons from a real deployment.
AI calls keep running but your CRM stops updating. Why webhook delivery fails quietly, how to catch it in a day, and a simple daily check that prevents it.
How a real deployment scores every failed AI call by drop reason and script stage, including bot suspicion, to know exactly which line to fix.
A real AI agent ended calls when callers said "thank you" or "no" mid-call. Why it happens, and the five-point closing checklist that fixed it.
Technical deep-dive into how voice AI handles code-switching between Hindi and English — the most common calling pattern in Indian business.
Technical deep-dive into Gemini Live 2.5 HD for voice AI — native audio processing, 30 HD voices, fast responses, and affective dialog.
Comparison of telephony providers for voice AI in India. Pricing, reliability, Indian number support, and integration complexity.
How to build an LLM-based post-call extraction pipeline. Architecture, template design, provider selection, and real-world results.
How AI detects caller sentiment in real-time and post-call. Cooperative, resistant, confused, hostile classification with real examples.
Developer guide to integrating voice AI via APIs, webhooks, and custom integrations. Real-time events, post-call data, and CRM connectivity.
Technical comparison of voice AI architectures: traditional STT-LLM-TTS pipeline vs native audio-to-audio. Latency, quality, and cost trade-offs.
Technical comparison of voicemail detection methods for AI voice agents. Prompt-based detection, pattern matching, and hybrid approaches.
Learn how we implemented a dynamic domain management system in Coolify that allows adding and removing multiple domains to applications without redeployment using Traefik's file provider for instant routing updates.
Learn how we built a comprehensive cross-posting automation system that publishes blog content to Dev.to, Medium, Hashnode, LinkedIn, and Hackernoon with a single command. Includes RSS feed generation, OAuth integration, and SEO-safe canonical URLs.
A straightforward guide to enabling multi-provider authentication in Next.js 15 with NextAuth.js.