Technical Lead, Real-Time Voice Translation

Quick Facts

• Role: Technical Lead for a real-time voice translation platform (mobile iOS/Android)
• Stage: 2-week discovery and technical architecture sprint
• Work model: 100% remote (Europe time zones)

Description

You will take technical ownership for the initial phase of an innovative mobile application that translates live conversations between two users on separate phones. The solution must achieve very low latency under real cellular network conditions while maintaining high translation quality and strict data privacy/compliance requirements. The project will develop for iOS and Android using React Native, native audio modules, a Python backend, and WebRTC streaming.

Responsibilities

• Design the overall solution architecture for real-time speech-to-speech translation, including comparing STT→MT→TTS vs. end-to-end approaches
• Architect streaming audio processing so translated speech can start playing before the full source utterance is complete
• Evaluate AI vendors for STT, MT, TTS, and WebRTC audio transport with emphasis on streaming and compliance needs
• Define and measure an end-to-end latency budget across the pipeline (including variability and tail latency)
• Evaluate TTS quality variants for naturalness, speaker similarity, streaming performance, multilingual coverage, and cost
• Define test methodology and quality criteria, including time-to-first-audio measurement using percentiles and tail analysis
• Prepare measurable Alpha-phase acceptance criteria and pass/fail thresholds agreed with the client
• Build a cost model for processing minutes across usage scenarios
• Assess feasibility and potential for on-device processing of selected pipeline components
• Set up technical project foundations: repository, CI/CD, application skeleton, and authentication basics
• Produce English technical architecture documentation (recommendations, justifications, comparisons, and decision records)
• Lead technical discussions with the client and justify decisions with data and analysis
• Plan the next stage scope for an Alpha build, estimate effort, and propose team structure and budget

Requirements

• Documented production experience in real-time audio processing systems, specifically streaming voice pipelines (not just batch transcription or basic request/response AI)
• Practical production experience in at least two of: Streaming STT; streaming/incremental MT; streaming TTS
• Production experience with WebRTC and understanding audio transmission behavior over unstable connections and cellular constraints
• Experience instrumenting and measuring latency on end devices (median and p95), analyzing latency distributions and extreme values
• Ability to assess technology providers against compliance requirements, including EU hosting and zero data retention
• Very good Python knowledge for backend development
• Familiarity with streaming pipeline orchestration frameworks (e.g., Pipecat or equivalents)
• Technical Lead/Architect experience including independent decision-making, documentation, and defense
• Ability to analyze trade-offs between latency, quality, cost, scalability, and data privacy
• Excellent English speaking and writing for client meetings and precise technical documentation
• Availability to start quickly and full commitment during the first two weeks (8 hours/day)

Benefits

• Fast start: collaboration can begin immediately (client may wait up to 2 weeks for onboarding)
• Stability: B2B contract for at least 3 months with possible long-term extension
• Fully remote work: no requirement for office/client visits
• Quick recruitment: one stage (technical interview with developers + HR) and feedback within ~1 week