Build Voice AI systems
Real-time STT, LLM and TTS pipelines, barge-in, interruption handling, answering-machine detection and structured post-call data.
VOICE AI / SIP / WEBRTC
Voice systems that work in production.
I design and build production-grade Voice AI and VoIP systems — from SIP and WebRTC infrastructure to real-time AI audio pipelines and high-concurrency calling platforms.
15+YEARS IN BACKEND & TELECOM
SIP · RTP · AIPRODUCTION EXPERTISE
E2EVOICE STACK OWNERSHIP
HOW I CAN HELP
The difficult part is rarely the API call.
I handle the architecture around the model: media flow, signalling, latency, concurrency, failure recovery and the browser experience.
Design telephony infrastructure
SIP routing, WebRTC, SBC architecture, media handling, NAT traversal, carrier integrations, redundancy and scaling.
Fix difficult voice problems
One-way audio, latency, dropped media, broken signalling and AI integrations that work in a demo but fail under real traffic.
SELECTED WORK
Representative
systems and
architecture.
High-concurrency Voice AI platform
Drachtio and FreeSWITCH media layer, Kubernetes deployment and a real-time speech pipeline designed for hundreds of simultaneous calls.
Browser-based outbound dialer
Embedded WebRTC softphone, early media, backend-controlled calls and voicemail handoff that releases the agent while playback continues.
Multi-protocol communications platform
OpenSIPS, RTPengine and Asterisk with secure SIP/WebRTC routing, backend event processing and production observability.
ABOUT
One engineer across the entire voice stack.
I’m a backend and VoIP engineer with 15+ years of experience building telecom infrastructure and real-time systems. I help teams move from a fragile proof of concept to a voice product they can operate, debug and scale.
FREESWITCH · ASTERISK · OPENSIPS · WEBRTCNODE.JS · RTPENGINE · KUBERNETES · REAL-TIME AI