Head of Research (AI)

Role overview

This is a senior leadership opportunity for a player-coach to head the research function for an organization building foundation models over relational data. The mission centers on scaling the scientific vision for graph-native models used in high-stakes enterprise prediction, while bridging the gap between cutting-edge research and production-grade systems that move customer metrics. The role blends people leadership, hands-on modeling, and tight partnership with data, product engineering, and go-to-market teams.

Responsibilities

- Own and evolve the research roadmap for foundation models focused on knowledge graphs, representation learning, and self-supervised or unsupervised methods. - Lead, mentor, and grow a team of research scientists and engineers, setting a high technical bar across the function. - Run a rigorous research cadence covering hypothesis definition, RFCs, disciplined experimentation, and data-driven decision-making. - Design and implement state-of-the-art graph neural networks for node, edge, and graph-level tasks, including multi-scale embeddings and temporal or inductive generalization. - Build reliable, reproducible training and evaluation pipelines using PyTorch and graph learning libraries, with distributed training and statistically sound benchmarks. - Partner with product engineering to productize models for batch and online inference, and with go-to-market teams to define customer success criteria and communicate impact to technical and executive stakeholders. - Account for real-world deployment challenges such as concept drift and requirements in regulated contexts.

Requirements

- 7+ years of experience in AI or machine learning, or a PhD plus 4 years of relevant industry experience. - Demonstrated record of shipping research into production: taking ideas from paper or prototype to scalable, reliable code that delivered measurable business impact. - Hands-on expertise in Python and PyTorch, with deep proficiency in graph learning libraries such as PyTorch Geometric or DGL. - Solid software engineering fundamentals, including testing, profiling, and building maintainable systems. - Experience mentoring technical contributors or managing a small team of roughly 2-6 scientists and engineers. - Professional working proficiency in both Portuguese and English.

Nice to have

- Experience with distributed training and inference at scale. - Deep familiarity with self-supervised or contrastive learning techniques applied to graphs. - Publication record at top-tier AI venues such as NeurIPS, ICML, or ICLR, or significant open-source contributions.