ML/AI Engineer — Real-Time Speech & Language Systems at BeyondNorme
Live audio is processed through a multi-stage intelligence system: transcription, sensitive data handling, structured extraction, decision routing, and continuous performance refinement.
The system operates on high-volume enterprise voice data, transforming unstructured conversation streams into structured, actionable outputs while ensuring sensitive information is removed prior to persistence. Outputs feed downstream systems that enable real-time operational use cases and iterative system improvement.
We are seeking the engineer responsible for owning the machine learning and data processing architecture, working directly with the CTO on system design decisions.
Remote-first. NDA required prior to full technical disclosure.
⸻
Work Experience: 4+ years professional experience
3+ years building production ML / AI systems
2+ years working with speech-to-text or ASR pipelines
⸻
Skills Required:
Python (FastAPI, async systems)
Speech-to-text / ASR systems (Whisper, Deepgram, AssemblyAI or equivalent)
LLM integration (Claude, GPT-4) and prompt design
PII detection and redaction systems (Presidio or equivalent frameworks)
Vector databases (pgvector, Qdrant, Pinecone)
Docker, Kubernetes, AWS
Model evaluation, monitoring, and performance optimization
⸻
Annual Salary Range: $140,000 – $200,000
(deferred until funding; contract rate equivalent: $70–$100/hr)
⸻
Equity: None (seats filled)
Funding Note:
“Compensation deferred until first institutional close. Founding team member — first to get paid when funding hits.”