Applied Scientist

About the Role

Before an agent can reliably operate software on a user's behalf, it needs to understand what should happen next and why: the underlying process, its steps, decision points and governing logic, branches, variations, dependencies, and long-tail edge cases.

In practice, this knowledge is rarely represented explicitly. It is distributed across messy event logs, unstructured documents, screen recordings, UI interaction traces, historical executions, and the tacit knowledge of domain experts.

As an Applied Scientist on this team, you will work across teams to develop process-agnostic, scalable learning systems that recover the latent structure, decision logic, and dynamics underlying complex process execution trajectories from sparse, noisy, highly correlated observations spanning substantial behavioral heterogeneity and organizational scale.

You will take heterogeneous data such as action logs, noisy screen traces, semi-structured documents, and variant-heavy workflows, and build systems that recover canonical process structure, meaningful variations, decision rules, dependencies, exceptions, and provenance.

The same observational data and learned representations also contribute directly to our work on computer-use models for last-mile interaction with enterprise software.

Responsibilities

Process discovery and representation

• Build pipelines that discover, extract, and formalize end-to-end business processes from event logs, UI interaction traces, screen recordings, and unstructured documents.
• Develop algorithms for process variant detection, noise removal, and trace consolidation.
• Design representations that preserve the information downstream agents need while collapsing irrelevant execution-level variation.
• Study the tradeoff between abstraction and fidelity.
• Build retrieval and grounding systems that incorporate SOPs, documents, and domain knowledge.

Evaluation and reliability

• Develop rigorous evaluation frameworks for process discovery and extraction, including LLM-as-Judge pipelines, conformance checking, and regression suites.
• Enumerate and systematically test exception paths, conditional branches, loops, rare variants, and failure modes.
• Make deliberate algorithmic tradeoffs so systems remain tractable across millions of traces and thousands of variants.
• Diagnose failures in production and turn recurring failure patterns into new algorithmic or research directions.

Collaboration

• Work closely with the computer-use agents team so the representations are robust enough for downstream autonomous execution.
• Work across research, engineering, and product to move promising methods from prototype to production.

Requirements

• Ambitious and Hungry.
• Problem Decomposition and Algorithmic Thinking, including graph algorithms, approximation, indexing, and time and space complexity.
• ML / DL Breadth across representation learning, sequence modeling, clustering, structured prediction, information retrieval, generative modeling, and foundation-model adaptation.
• Cross-Domain Research Translation.
• Research Judgment.
• Applied Foundation Model Systems, including retrieval, structured generation, evaluation, grounding, and model adaptation.

Preferred Experience

• A strong research track record, demonstrated through publications in top-tier machine learning, artificial intelligence, reinforcement learning, natural language processing, computer vision, or data mining venues.
• Process mining: event log analysis, process discovery algorithms, conformance checking, or variant analysis.
• Information extraction from documents: NER, relation extraction, table extraction, or structured output generation.
• Multimodal data processing from screen recordings, UI screenshots, or interaction traces.
• Foundation-model adaptation, including supervised fine-tuning, LoRA, synthetic-data methods, distillation, and reinforcement-learning-based post-training.

Qualifications

• Strong software engineering in Python, experience with ML frameworks such as PyTorch, and the ability to write clean, tested, production-quality code.
• Communication: you can explain algorithmic tradeoffs to researchers and product managers, and write clear technical and design documents.