AI Research Engineer, Model Compression and Quantization at Tether

Confirmed open

Compress language and vision-language models using quantization, distillation and pruning for efficient edge deployment. The employer lists Remote — Brazil.

Tether

Location: Remote — Brazil
Employment: fulltime_permanent
Work arrangement: remote
Technologies: C++, PyTorch

Role overview

Tether’s AI research team works to reduce the size, latency and compute cost of language and multimodal models while protecting output quality. This role researches compression for resource-limited devices, including smartphones. Each requisition specifies its own remote location.

Responsibilities

Apply low-bit and mixed-precision quantization to reduce model footprint and inference latency while measuring quality.
Distil capabilities from large teacher models into smaller models for text, image and audio tasks.
Prune redundant parameters and attention heads and compare effects on accuracy, memory and throughput.
Design reproducible compression experiments, document methods and results, and collaborate on technical findings.
Research emerging compression approaches and publish relevant work in AI research venues.

Qualifications

Computer science or related degree; a PhD in NLP or machine learning and research publications are preferred.
Experience with PyTorch or equivalent deep-learning frameworks and model architectures such as transformers.
Practical quantization work using quantization-aware training and post-training quantization.
Research and implementation experience with knowledge distillation and model pruning.
Understanding of training, backpropagation, optimisation and fine-tuning; C++ for low-level kernels is a plus.

About Tether

Tether develops digital-token products, including USDT, alongside projects in peer-to-peer communications and AI. The employer describes a globally distributed remote team.

How to apply

Review the original posting and apply through the employer’s careers page.

View source and apply Apply directly

Source checked: 2026-09-29

Show what you can do

Build a free profile around your CV and real work. Share its link when you are ready. Applications for this role still go directly to the employer.

Build a free profile