Serve as the technical reference for ML/AI and data within the Investment Products Community, influencing architecture standards and product decisions across the company.
Define the technical vision and roadmap for AI/ML products and solutions, prioritizing by impact, risk, and feasibility, in alignment with key stakeholders.
Design and implement data and ML pipelines (training, evaluation, inference) with a focus on reliability, performance, and observability.
Evolve agent and multi-agent architectures (orchestration, memory, tools, guardrails, and security), balancing cost, latency, and quality.
Implement guardrails and security and privacy best practices (LGPD — Brazilian Data Protection Law, DLP, access control, mitigation of prompt injection).
Conduct trade-off analyses and architecture decisions (build vs. buy, platform vs. service) with a FinOps and scalability perspective.

Strong applied Data Science experience with ML/AI delivered to production and measurable impact.
Proficiency in Python and data/ML libraries (Pandas, NumPy, scikit-learn; PyTorch/TensorFlow desirable).
Code versioning (Git/GitHub) and engineering best practices (code review, CI/CD).
Data modeling and data architecture; advanced SQL; experience with relational and non-relational databases.
Experience with Big Data and distributed processing (Spark) and batch/streaming pipelines.
MLOps/LLMOps, experiment tracking, model registry, observability, monitoring, and drift detection.
Advanced cloud knowledge (AWS preferred: S3, Lambda, Glue; plus: Step Functions, ECS/EKS, SageMaker).
RAG and vector databases (pgvector, OpenSearch, Pinecone, FAISS), LLM/agent evaluation, prompting techniques, caching, and optimization.
Big Data/Analytics ecosystem (Databricks/EMR, Kafka/Kinesis) and FinOps for AI workloads.
Experience developing agentic solutions for processing large volumes of data.
Experience with .NET Core.

Data Science Specialist – Machine Learning

Key skills