Senior Causal Discovery & Differential Privacy Engineer

corpora-jobs-1778796293285-db9d41c6 - Frontier Development

Research EngineerSenior1 position

⚡

Why This Role is Different

Frontier Development Role

You’ll pioneer a privacy‑aware causal discovery engine that powers adversarially robust counterfactual explanations. Your work will sit at the intersection of causal inference, differential privacy, and adversarial machine learning, enabling trustworthy explanations in multi‑modal, multi‑agent systems.

The Frontier Element

Building a causal graph that is both statistically sound and privacy‑preserving in high‑dimensional multimodal settings pushes the boundary of what causal discovery can achieve in production adversarial environments.

🔬

Project Context

Research Area

Causal Graph Learning for Adversarial Steering

From: Counterfactual Explanation Robustness to Adversarial Noise

Why This Role is Critical

The FCA depends on a high‑fidelity, privacy‑preserving causal graph to steer perturbations along semantically valid edges; without it the counterfactuals become spurious and trust erodes.

What You Will Build

A scalable causal discovery engine that learns directed acyclic graphs from multimodal data, embeds differential privacy guarantees, and exposes a causal API for downstream modules.

🛠

Key Responsibilities

Design and implement fast causal discovery algorithms (e.g., FCI, GAC) that scale to millions of features across vision, language, and graph modalities.
Integrate differential privacy mechanisms (e.g., DP‑SGD, Laplace noise) into the discovery pipeline to protect sensitive data while maintaining causal accuracy.
Develop a causal API that exposes intervention semantics and edge‑confidence scores to downstream modules (CECAS, MARM).
Validate causal consistency using synthetic and real‑world datasets, iterating on graph refinement heuristics.
Collaborate with the Diffusion Model Architect to embed causal constraints into the diffusion guidance process.

🎯

Required Skills & Experience

Technical Must-Haves

Causal inference and graphical models

Expert

Designing and evaluating DAGs for high‑dimensional multimodal data.

Differential privacy

Advanced

Implementing DP‑SGD, Rényi DP, and privacy accounting for causal discovery.

Python, PyTorch, and graph libraries (NetworkX, DGL)

Expert

Building scalable pipelines and graph neural network components.

Statistical testing and causal validation

Advanced

Assessing faithfulness, identifiability, and robustness of learned graphs.

Experience Requirements

5+ years in causal discovery or related fields (e.g., structural causal models, causal inference in ML).
Published research on causal graph learning or privacy‑preserving analytics.
Hands‑on experience building production‑grade pipelines for large‑scale data.

Education

PhD in Statistics, Computer Science, or a related field with a focus on causal inference.

⭐

Preferred Skills

Experience with federated or vertical learning environments.
Knowledge of graph‑aware diffusion models.
Familiarity with healthcare or finance regulatory constraints.

🤝

You Will Thrive Here If...

Thrives in high‑autonomy settings and can iterate rapidly from theory to prototype.
Shows a strong bias toward action, delivering working code over theoretical papers.

📈

Impact & Growth

12-Month Impact

Within 12 months, deliver a causal discovery service that reduces spurious counterfactuals by 90%, enabling downstream modules to generate actionable explanations that remain valid under adversarial perturbations.

Growth Opportunity

Lead a dedicated causal inference team, shaping the next generation of privacy‑preserving, causally grounded AI systems.

Ready to Push the Boundaries?

If this sounds like the challenge you have been looking for, we want to hear from you. We value what you can build over where you have been.