Baseten
Software Engineer - Voice AI (Inference Runtime)
$165K–$330K
San Francisco +4 · Hybrid · Full-time · Posted Apr 23, 2026
- Location
- San Francisco
Toronto
New York
Montreal
Seattle - Education
- Bachelor's degree or higher in Computer Science or related field
About the job
Own Baseten’s Voice AI inference stack, building and operating real-time model-serving systems for speech and voice-agent workloads.
What you'll do
- Lead Voice AI product areas from architecture and implementation through rollout and production operations.
- Build high-performance, large-scale serving systems for STT, TTS, and voice agents, improving latency, throughput, and GPU efficiency.
- Coordinate delivery across engineering teams and mentor teammates through reviews, design docs, and technical leadership.
What they ask for
- Bachelor's degree or higher in Computer Science or a related field.
- Proven track record owning production-grade, real-time, large-scale systems where tail latency (p99) matters.
- Proficient in one or more popular programming or scripting languages; Python proficiency is a plus.
- Comfortable using AI coding assistants as a daily productivity multiplier.
Nice to have
- Experience with pipeline-level model runtime optimizations or developer platforms for ML or infrastructure products.
- Familiarity with speech/audio ML models, model-serving runtimes, GPU performance profiling, or customer-facing engineering.
Pay
$165K – $330K USD base salary
Tools
Python, Docker, Kubernetes, vLLM, TensorRT, ONNX, PyTorch Profiler, Claude Code
AI focus
Develop and operate inference infrastructure for STT, TTS, and voice-agent models, including runtime optimization and voice-model customization workflows.
Model-written summary · the original posting is authoritative