---
title: "Senior Real-Time Voice Agent & Streaming Systems Engineer"
slug: "senior-real-time-voice-agent-streaming-systems"
id: "818aaac8-a1bf-4985-9edd-4753a310c8bf"
department: "Applied AI"
department_code: "APP"
location: "London, UK"
workplace_type: "Hybrid"
employment_type: "Full-time"
compensation: "£105k – £145k (annual package)"
is_remote: false
is_featured: false
headcount: 2
canonical_url: "https://jobs.talent247.co/jobs/senior-real-time-voice-agent-streaming-systems"
markdown_url: "https://jobs.talent247.co/jobs/senior-real-time-voice-agent-streaming-systems.md"
apply_endpoint: "https://jobs.talent247.co/api/public/jobs/senior-real-time-voice-agent-streaming-systems/apply"
schema_endpoint: "https://jobs.talent247.co/api/public/jobs/senior-real-time-voice-agent-streaming-systems/form-schema"
webmcp_tool: "apply_for_role"
updated_at: "2026-11-15T17:42:55.637Z"
content_signal: "search=yes, ai-input=yes, ai-train=no"
---

# Senior Real-Time Voice Agent & Streaming Systems Engineer

## Position Telemetry

- **Reference:** `SENIOR-REAL-TIME-V`
- **Practice / Sector:** Applied AI (`APP`)
- **Location:** London, UK
- **Workplace Type:** Hybrid
- **Employment Type:** Full-time
- **Compensation:** £105k – £145k (annual package)
- **Headcount:** 2 openings
- **Canonical HTML URL:** https://jobs.talent247.co/jobs/senior-real-time-voice-agent-streaming-systems

## Role Dossier & Specification

THE CLIENT & MANDATE
Our client is a globally renowned generative audio and voice AI unicorn headquartered in London and New York, powering real-time conversational voice agents for technology platforms, healthcare providers, financial services, and interactive entertainment worldwide. In live duplex voice interaction, every 20 milliseconds of network jitter, turn-detection delay, or tool-execution lag separates a magical human-like conversation from an awkward robotic pause.

They have engaged Talent247 to hire two Senior Real-Time Voice Agent & Streaming Systems Engineers in London. Joining their Conversational Agents Infrastructure group, the appointed engineers will architect the globally distributed WebRTC/SIP media plane and stateful streaming orchestration engine coordinating sub-200ms speech recognition, LLM reasoning, MCP/webhook tool calls, and expressive neural audio synthesis across hundreds of thousands of concurrent calls.

KEY RESPONSIBILITIES
• Architect, scale, and operate globally distributed, edge-routed WebRTC and SIP telephony media servers in Rust, Go, and Python delivering ultra-low-jitter bidirectional audio streaming.
• Engineer stateful real-time conversation loops handling semantic turn-end prediction, natural back-channeling, mid-sentence user barge-in cancellation, and dynamic background-noise suppression.
• Build deterministic asynchronous tool-execution brokers (supporting webhooks, client-side SDK callbacks, and Model Context Protocol servers) that let agents fetch data or execute actions mid-call while gracefully filler-bridging or streaming interim speech.
• Instrument millisecond-level end-to-end latency waterfalls (network RTT, VAD/turn-taking, ASR, LLM TTFT, TTS first-chunk) across global regions using OpenTelemetry and ClickHouse.
• Design chaos-engineering, call-replay debugging, and high-concurrency load-testing harnesses ensuring 99.99% uptime under sudden enterprise call spikes.
• Partner directly with audio ML researchers to co-design native speech-to-speech streaming protocols replacing cascaded text bottlenecks.

TECHNICAL ENVIRONMENT & OFFERING
Their systems stack runs on Rust, Go, Python, WebRTC, Pion/LiveKit/custom RTP media engines, SIP trunks, Kubernetes, ClickHouse, Redis, and GPU inference clusters. Working hybrid in central London, our client offers £105,000–£145,000 GBP base salary plus substantial unicorn equity, private healthcare, and UK visa/relocation sponsorship.

## Core Requirements

ESSENTIAL QUALIFICATIONS
• 5+ years of backend and real-time distributed systems engineering experience building latency-critical streaming media, telephony, gaming, or real-time AI platforms.
• Strong systems programming skills in Rust or Go alongside Python, with deep mastery of asynchronous concurrency, lock-free audio ring buffers, and network tail-latency tuning.
• Direct production experience with WebRTC (ICE/STUN/TURN, Opus/PCM codecs, jitter buffers), WebSockets, gRPC bidirectional streaming, and/or SIP telephony infrastructure.
• Experience integrating real-time ML inference streams (streaming ASR, LLM token streams, chunked TTS, or neural VAD) under sub-300ms round-trip budgets.
• High ownership mentality in a fast-shipping engineering culture where engineers own architecture from RFC through global production rollout.

PREFERRED / BONUS EXPERIENCE
• Experience building conversational AI voice agents, CCaaS telephony bridges, or Model Context Protocol (MCP) tool runtimes.
• Familiarity with embedded audio DSP (echo cancellation, noise suppression) or global Anycast/edge routing.

## Application Schema & Direct Apply (`apply_for_role`)

Candidates and authorized AI agents on behalf of a candidate can submit an application via the declarative WebMCP form tool `apply_for_role` on https://jobs.talent247.co/jobs/senior-real-time-voice-agent-streaming-systems or via HTTP `POST https://jobs.talent247.co/api/public/jobs/senior-real-time-voice-agent-streaming-systems/apply` (`application/json` or `multipart/form-data`). Non-browser HTTP agents must include header `X-Talent247-Intake-Token` obtained from `GET https://jobs.talent247.co/api/public/intake-token` (or automatically returned in `GET https://jobs.talent247.co/api/public/jobs/senior-real-time-voice-agent-streaming-systems/form-schema`).

### Form Fields

- `full_name` (text, required): Full name
- `email` (email, required): Email
- `phone` (phone, optional): Phone
- `location` (text, optional): Location
- `linkedin_url` (url, optional): LinkedIn URL
- `cover_letter` (textarea, optional): Cover letter
- `consent` (checkbox, required): I consent to my data being processed for this application
- `cv_file` / `cv_base64` + `cv_filename`: Upload a CV in .pdf or .docx format (<= 10 MB) or provide an `https://` LinkedIn profile URL in `linkedin_url`.

### Alternative: Talent Pool Intake (`join_talent_pool`)

If this mandate is not an exact match, register once in the Talent247 Talent Pool at https://jobs.talent247.co/talent-pool (`https://jobs.talent247.co/talent-pool.md`) via WebMCP tool `join_talent_pool` or `POST https://jobs.talent247.co/api/public/talent-pool`.
