Remote Senior ML Engineer: ASR/LLMs & AWS MLOps

Outlier; Quilmes; Remote; ARS 139,086,000 - 216,357,000; Full time

Job summary

Outlier is seeking an Applied ML Engineer (Senior) to lead the development and deployment of speech-to-text models for Brazilian Portuguese and Latin American Spanish, while contributing to LLM fine-tuning and medical classifiers.

This hands-on role focuses on taking models from experimentation to production, building scalable pipelines, and ensuring reliability for real-world AI products in a fast-paced environment.

Qualifications

• Strong experience building and operationalizing ML models in production.
• Hands-on experience with ASR / speech models / audio ML.
• Proficiency in PyTorch or modern ML frameworks.
• Experience with ML infrastructure and distributed training.
• Strong Python programming skills.
• Ability to work independently in ambiguous environments.
• Experience collaborating across technical and non-technical teams.

Responsibilities

• Design, train, fine‑tune, and deploy ML models for speech and language applications
• Build and maintain training pipelines and evaluation frameworks
• Work on ASR systems for domain‑specific medical conversations
• Develop and improve data pipelines and ML infrastructure
• Collaborate on LLM fine‑tuning and classifier development
• Optimize model performance, reliability, and scalability
• Partner with product and domain teams to translate business needs into ML solutions
• Contribute to production deployment and monitoring of ML systems

Skills

ML in production

ASR / speech ML

Python

LLMs / fine-tuning

Data pipelines

Collaboration

Ambiguity handling

Model deployment

Tools

PyTorch

Orchestration frameworks

Experiment tracking

GPU orchestration

Job description

Outlier is seeking an Applied ML Engineer (Senior) to lead the development and deployment of speech-to-text models for Brazilian Portuguese and Latin American Spanish, while contributing to LLM fine-tuning and medical classifiers.

This hands-on role focuses on taking models from experimentation to production, building scalable pipelines, and ensuring reliability for real-world AI products in a fast-paced environment.