Full-Stack Software Engineer – AI Voice

About the role

We are seeking a senior Full‑Stack Software Engineer to own and improve the end‑to‑end experience of AI‑powered voice interviews. You will work across backend services, real‑time voice agents, infrastructure, observability, and evaluation pipelines to ensure high quality, reliability, and performance.

Key responsibilities

• Develop a deep understanding of the complete voice‑interview lifecycle, from participant admission to transcription, recording, and analysis.
• Design repeatable evaluation frameworks for prompts, models, voice pipelines, providers, and infrastructure changes.
• Define, collect, and analyze quantitative metrics such as latency, error rates, transcript accuracy, and resource consumption.
• Evaluate qualitative aspects like question relevance, conversational coherence, and empathy.
• Build automated evaluation tools using transcripts, telemetry, recordings, and LLM‑as‑judge approaches.
• Diagnose issues across application code, external AI providers, media infrastructure, and observability data.

Required profile

• Strong software engineering experience with Python and/or TypeScript.
• Background in building, testing, or evaluating LLM‑powered applications.
• Understanding of real‑time voice systems (STT, TTS, VAD, endpointing, interruption handling).
• Analytical mindset with experience turning product‑quality questions into measurable metrics.
• Ability to write automated tests and maintain production‑grade systems.
• Excellent written communication and documentation skills.

Required skills

• Python
• TypeScript
• Node.js
• OpenAI
• Anthropic
• OpenSearch
• pgvector
• Pinecone
• Terraform
• CDK
• LiveKit
• WebRTC
• Langfuse
• OpenTelemetry