Engineering
conversational AI.
From the first customer requirement to the last audible word. Twelve detailed articles drawing on Bolna and Pipecat delivery, speech-model experiments and the systems around a real telephone call.
Conversational AI Design: the call is the product
From customer requirements to an audio, reasoning and business-action system.
Why an 800 ms voice response is a systems problem
An explicit latency budget, turn detection, cancellation and playback correctness.
A Bolna-based telephony agent, from connection to hangup
Call lifecycle, asynchronous queues, provider adapters and production boundaries.
Designing a configurable voice platform with Pipecat
Frames, lifecycle, SEDA principles and interoperable cascade and native-audio paths.
Choosing the speech and reasoning models for a voice agent
Provider trade-offs, native audio, language coverage and a workload-specific Pareto frontier.
Self-hosted Whisper: streaming, scheduling and the concurrency trap
Rolling recognition, stable prefixes, GPU queues and a capacity calculation for real calls.
Adapting a 2026 speech recognizer for English, Hindi, Singlish and Hinglish
A Qwen3-ASR reference experiment, plus an honest analysis of a completed Whisper Tiny run.
Speech synthesis that can survive a real customer call
Pronunciation, phrase streaming, a self-hosted Indic TTS candidate and adaptation protocol.
Context, memory and tools: where a voice agent becomes a business system
Conversation state, retrieval, streaming function calls, idempotency and interruption-safe actions.
Running voice agents on AWS and GCP without losing the conversation
Cloud topology, Bedrock, GKE/EKS, GPU serving, admission control and graceful draining.
Evaluating a voice agent: listen to the call, inspect the action
A practical evaluation system across audio, interaction, policy and verified business outcomes.
What a voice agent changes for the business
Unit economics and business outcomes from the client platform and Zoice.