Hire · Haris Ullah Khan

Hire a Voice AI Developer

Haris Ullah Khan is a voice AI developer you can hire to ship agentic calling agents on both paths - chained STT→LLM→TTS pipelines for max control and tooling/MCP, or speech-to-speech (Hume, Cartesia, Grok Voice) when latency matters most - plus LiveKit, Twilio, and production observability.

Built and led voice/agent platforms handling thousands of monthly AI calls. I set up agentic voice AI with chained pipelines or speech-to-speech, then wire CRM, transfers, and evals. Available remotely at $20.00/hr.

RoleFull Stack AI Developer
Experience6+ years
Rate$20.00/hr
LocationLahore, Pakistan · Remote
01

What you get

Chained or speech-to-speech

STS (Hume, Cartesia, Grok Voice) wins on latency; chained STT→LLM→TTS wins on per-layer control, tooling, and MCP. I ship both and help you pick from product goals.

Real-time voice infrastructure

PSTN/SIP, low-latency streaming, live transfers, and durable workflows so agents stay reliable under real call volume.

Full product surface

Agent studio, dashboards, call logs, CRM hooks, and APIs - the SaaS layer around the voice pipeline.

02

Proof at a glance

  • AI employee / voice platforms with thousands of monthly calls
  • Chained STT-LLM-TTS and speech-to-speech (Hume) in production
  • LiveKit, Twilio, Deepgram, ElevenLabs, Temporal in production stacks
  • HVAC calling agent and multi-channel conversational platforms shipped
03

FAQ

What is a voice AI developer?

A voice AI developer builds systems that talk to users over phone or web audio in real time - speech-to-text, LLM reasoning, text-to-speech, speech-to-speech, and telephony. Haris Ullah Khan ships those pipelines with LiveKit, Twilio, and production monitoring.

Do you build chained pipelines or speech-to-speech?

Both. Speech-to-speech (Hume, Cartesia, Grok Voice) is more latency-friendly for natural conversation, but tooling and MCP support are harder. A chained STT→LLM→TTS pipeline gives more control over each layer and stronger tool wiring, but needs fine-tuning at every stage. I recommend from your latency budget and tool requirements.

Do you work with Retell, Vapi, or custom stacks?

Yes. Experience includes Retell/Make-style agent builds and custom stacks (LiveKit + STT/LLM/TTS or speech-to-speech providers) when you need more control, cost efficiency, or deeper product integration.

How fast can a calling agent go live?

A focused MVP can ship in weeks when scope is clear (one flow, one language, one CRM). Complex multi-agent platforms take longer - scoped after a discovery call.

Can you improve an existing voice agent?

Yes - latency, interruption handling, tool calling, evaluations, and transfer logic are common engagements for teams that already have a prototype.

04

Explore next

Ready to scope a build? Book a free intro. No pressure, clear next steps.