Real-Time Speech ML Engineer — Sub-Second Latency

Sellsig; Minnesota; On-site; USD 100,000 - 130,000; Full time

Job summary

Sellsig is seeking an experienced ML Engineer to build real-time speech and language processing pipelines. The role focuses on developing streaming ASR and objection detection systems.

With at least 4 years of experience in applied ML, candidates should excel in Python and have practical exposure to low-latency implementations. The position emphasizes driving end-to-end latency and integrating modern LLM tools.

Qualifications

• 4+ years in ML/applied ML, with production work in speech, NLP, or real-time inference.
• Experience deploying models in low-latency, streaming settings.
• Comfortable owning ambiguous, research-flavored problems end to end.

Responsibilities

• Build and tune streaming ASR and intent/objection-detection pipelines.
• Drive down end-to-end latency while holding accuracy.
• Evaluate and integrate speech and LLM providers.

Skills

Machine Learning

Python

ASR/diarization

Low-latency streaming

Evaluation metrics

WebRTC

Tools

Deepgram

AssemblyAI

Job description

Sellsig is seeking an experienced ML Engineer to build real-time speech and language processing pipelines. The role focuses on developing streaming ASR and objection detection systems.

With at least 4 years of experience in applied ML, candidates should excel in Python and have practical exposure to low-latency implementations. The position emphasizes driving end-to-end latency and integrating modern LLM tools.