Head of Research (AI)

BrazilRemoteFull-timeexecutive

OtherResearch

2 views0 saves0 applied

Apply Now

Quick Summary

Requirements Summary

7+ years of professional experience in AI/ML, or a PhD combined with at least 4 years of relevant experience. Demonstrated ability to take research concepts, academic papers,

Technical Tools

OtherResearch

This is a hands-on scientific leadership role focused on advancing foundation models for relational and graph-structured data.
You will define the research vision, raise the technical bar, and lead work that turns cutting-edge AI research into production-ready capabilities.
The role spans graph learning, representation learning, self-supervised methods, large-scale modeling, and temporal generalization.
You will combine deep technical expertise with people leadership, mentoring research scientists and engineers while shaping a rigorous research culture.
Your work will directly support high-stakes enterprise applications where relationships between entities are critical to accurate decision-making.
You will collaborate closely with data, engineering, MLOps, product, and commercial teams to move research from hypotheses and prototypes into measurable business impact.
The environment values scientific rigor, experimentation, technical ownership, reproducibility, and research that delivers meaningful results in production.

Own and evolve the research roadmap for foundation models focused on relational data, knowledge graphs, representation learning, and self-supervised or unsupervised approaches.
Define the scientific strategy and research priorities, aligning technical exploration with production and business objectives.
Lead, mentor, and grow a high-performing team of research scientists and engineers.
Establish a rigorous research process covering hypothesis definition, RFCs, experimentation, evaluation, documentation, and data-driven decision-making.
Design and implement state-of-the-art Graph Neural Networks for large-scale relational datasets.
Solve complex node-, edge-, and graph-level learning problems, including multi-scale embeddings and temporal or inductive generalization.
Build reliable and reproducible training and evaluation pipelines using Python, PyTorch, PyTorch Geometric, and distributed training technologies.
Define and maintain high-quality benchmarks and evaluation methodologies to ensure statistically rigorous model comparisons.
Partner with product engineering and MLOps teams to transition research models into scalable, reliable batch and online inference systems.
Collaborate with commercial and go-to-market teams to define measurable success criteria for enterprise applications and communicate technical impact to both technical and executive audiences.
Account for real-world production challenges such as concept drift, model reliability, scalability, and deployment within regulated or risk-sensitive environments.
Provide technical direction across research and engineering initiatives while remaining actively involved in hands-on research and development.

Requirements

~1 min read

7+ years of professional experience in AI/ML, or a PhD combined with at least 4 years of relevant experience.
Demonstrated ability to take research concepts, academic papers, or prototypes through to scalable, reliable production systems that deliver measurable business impact.
Advanced hands-on proficiency in Python and PyTorch.
Deep experience with graph learning frameworks such as PyTorch Geometric or DGL.
Strong software engineering fundamentals, including testing, profiling, reproducibility, maintainability, and production-quality development.
Proven experience leading technical projects, mentoring researchers and engineers, or managing a small technical team of approximately 2–6 people.
Strong understanding of graph-based modeling and relational data.
Experience with self-supervised or contrastive learning techniques, particularly for graph-based applications, is highly desirable.
Experience with distributed model training and inference is a plus.
Strong scientific communication and ability to translate complex research into clear technical and business outcomes.
Professional proficiency in both Portuguese and English.
A strong publication record at leading AI conferences such as NeurIPS, ICML, or ICLR, or significant open-source contributions, is considered an advantage.