AI engineer jobs in Brazil on Rex.zone focus on building and deploying machine learning systems that power real-world AI products, including LLM training pipelines, RLHF workflows, evaluation harnesses, and data-centric quality improvements. You will collaborate remotely across teams to design model architectures, optimize training and inference, and improve training data quality through labeling strategy, QA evaluation, prompt evaluation, and safety checks. This page helps you explore Remote, Full-Time roles and understand what AI engineers do across NLP, computer vision, and content safety domains—then apply through Rex.zone.
Title: AI Engineer Jobs in Brazil Date: 25-02-2026 Company: Rexzone Country: US Remote Type: Remote Employment Type: FULL_TIME Experience Level: Mid-Senior Industry: Technology Job Function: Engineering Skills: AI engineering, machine learning, deep learning, LLM training pipelines, RLHF, prompt evaluation, model evaluation, MLOps, Python, PyTorch Salary Currency: USD Salary Min: 63360 Salary Max: 126720 Pay Period: YEAR
As an AI Engineer, you will design, train, evaluate, and deploy ML models with a strong emphasis on LLM training pipelines and measurable model performance improvement. You will build data and evaluation workflows that connect model development to training data quality, annotation guidelines compliance, and QA evaluation. Typical work includes training and fine-tuning models, implementing RLHF or preference learning loops, running prompt evaluation, designing automatic and human-in-the-loop model evaluation, and partnering with data operations to improve labeled datasets (NER, content safety labeling, and computer vision annotation when relevant).
You will: (1) build and iterate on ML/LLM systems from prototype to production, (2) develop evaluation harnesses for offline and online testing (including prompt evaluation and LLM evaluation), (3) implement RLHF-style feedback loops using human preference data when required, (4) improve training data quality by defining labeling strategy, validation checks, and annotation guidelines, (5) partner with QA evaluation and data labeling teams to reduce noise and bias, (6) optimize inference latency and cost, and (7) document experiments, metrics, and rollouts.
Common workflows include: supervised fine-tuning, instruction tuning, retrieval-augmented generation (RAG), preference modeling for RLHF, reward modeling, prompt evaluation, red-teaming and content safety labeling, named entity recognition (NER) dataset development, computer vision annotation strategy, dataset curation, and continuous evaluation with A/B testing and regression suites.
Mid-Senior experience building ML systems; strong Python engineering; hands-on deep learning with PyTorch or equivalent; experience with model evaluation and metrics; ability to reason about training data quality and annotation guidelines compliance; familiarity with LLM training pipelines and/or NLP systems; comfort collaborating remotely across product, engineering, and data operations.
Experience with RLHF, preference data, or reward modeling; experience with prompt evaluation and LLM evaluation frameworks; MLOps experience (CI/CD for ML, model registry, monitoring); experience with data labeling and QA evaluation processes; familiarity with content safety labeling, NER, or computer vision annotation; experience deploying models with scalable inference and observability.
Typical stack: Python, PyTorch, CUDA (optional), Hugging Face, vector databases for RAG, experiment tracking, containerization, cloud compute, and evaluation toolchains. You may also interface with annotation platforms, QA sampling plans, and dataset versioning to improve training data quality over time.
This role is Remote and full-time. Collaboration includes async documentation, clear experiment writeups, and structured evaluation reviews. You will work cross-functionally with engineering, product, and data teams to translate model requirements into datasets, labeling instructions, QA evaluation criteria, and measurable success metrics.
Apply via Rex.zone by submitting your resume and a brief overview of ML systems you have built (training, evaluation, deployment). Highlight experience with LLM training pipelines, RLHF or preference learning, prompt evaluation, and any work improving training data quality through data labeling and QA evaluation.
AI Engineers build ML systems end-to-end: data and training pipelines, model development, evaluation harnesses, and production deployment. On Rex.zone, many roles also touch LLM training pipelines, prompt evaluation, RLHF-style feedback loops, and training data quality improvements via labeling and QA evaluation.
Yes. The job metadata for this posting is explicitly Remote and FULL_TIME.
Common domains include NLP, LLM applications, retrieval-augmented generation, content safety, and sometimes computer vision. Work often includes model evaluation, prompt evaluation, and dataset improvements such as NER labeling and QA evaluation sampling.
Not always, but RLHF, preference learning, and reward modeling are strong advantages for LLM-focused roles. Equivalent experience with human-in-the-loop model evaluation, prompt evaluation, and data-centric iteration can also be relevant.
AI engineering outcomes depend on training data quality. AI Engineers often define labeling strategy, write annotation guidelines, partner with labeling vendors, and set up QA evaluation to ensure annotation guidelines compliance and improve model performance.
Include examples of ML systems you shipped, your approach to model evaluation, familiarity with LLM training pipelines, MLOps practices, and any experience with RLHF, prompt evaluation, data labeling, QA evaluation, NER, computer vision annotation, or content safety labeling.
'%20stroke-width='1.1184'/%3e%3cline%20x1='13.9799'%20y1='14.5392'%20x2='13.9799'%20y2='0.000106738'%20stroke='url(%23paint1_linear_5108_6545)'%20stroke-width='1.1184'/%3e%3cline%20y1='-0.559198'%20x2='29.0783'%20y2='-0.559198'%20transform='matrix(-1%208.74228e-08%208.74228e-08%201%2029.0781%20136.539)'%20stroke='url(%23paint2_linear_5108_6545)'%20stroke-width='1.1184'/%3e%3cline%20y1='-0.559198'%20x2='14.5391'%20y2='-0.559198'%20transform='matrix(1.31134e-07%201%201%20-1.31134e-07%2014.5391%20122)'%20stroke='url(%23paint3_linear_5108_6545)'%20stroke-width='1.1184'/%3e%3cdefs%3e%3clinearGradient%20id='paint0_linear_5108_6545'%20x1='29.0781'%20y1='-0.499878'%20x2='-0.000152544'%20y2='-0.49988'%20gradientUnits='userSpaceOnUse'%3e%3cstop%20stop-color='%23666666'/%3e%3cstop%20offset='0.336538'%20stop-color='white'/%3e%3cstop%20offset='0.495192'%20stop-color='white'/%3e%3cstop%20offset='0.653846'%20stop-color='white'/%3e%3cstop%20offset='1'%20stop-color='%23666666'/%3e%3c/linearGradient%3e%3clinearGradient%20id='paint1_linear_5108_6545'%20x1='15.0391'%20y1='14.5392'%20x2='15.0391'%20y2='0.000106877'%20gradientUnits='userSpaceOnUse'%3e%3cstop%20stop-color='%23585858'/%3e%3cstop%20offset='0.745192'%20stop-color='white'/%3e%3cstop%20offset='1'%20stop-color='white'/%3e%3c/linearGradient%3e%3clinearGradient%20id='paint2_linear_5108_6545'%20x1='0'%20y1='0.5'%20x2='29.0783'%20y2='0.5'%20gradientUnits='userSpaceOnUse'%3e%3cstop%20stop-color='%23666666'/%3e%3cstop%20offset='0.336538'%20stop-color='white'/%3e%3cstop%20offset='0.495192'%20stop-color='white'/%3e%3cstop%20offset='0.653846'%20stop-color='white'/%3e%3cstop%20offset='1'%20stop-color='%23666666'/%3e%3c/linearGradient%3e%3clinearGradient%20id='paint3_linear_5108_6545'%20x1='0'%20y1='0.5'%20x2='14.5391'%20y2='0.5'%20gradientUnits='userSpaceOnUse'%3e%3cstop%20stop-color='%23585858'/%3e%3cstop%20offset='0.745192'%20stop-color='white'/%3e%3cstop%20offset='1'%20stop-color='white'/%3e%3c/linearGradient%3e%3c/defs%3e%3c/svg%3e)
We believe exceptional intelligence deserves exceptional pay. Our platform consistently offers rates above the industry average, rewarding experts for their true value and real impact on frontier AI. Here, your expertise isn't just appreciated—it's properly compensated.
No office. No commute. No constraints. Our fully remote workflow gives experts complete flexibility to work at their own pace, from any country, any time zone. You focus on meaningful tasks—we handle the rest.
AI trainers are the heart of our company. We treat every expert with trust, humanity, and genuine appreciation. From personalized support to transparent communication, we build long-term relationships rooted in respect and care.
Ready to Shape the Future of Engineering?
Apply Now.