The Role
Lead research and production development of scalable AI systems that infer business-process structure from event logs, UI traces, recordings, and documents. Develop process discovery, representation learning, retrieval, evaluation, grounding, and foundation-model adaptation methods. Establish reliability standards, diagnose production failures, guide architecture, collaborate across research and engineering, and mentor scientists and engineers. The role emphasizes algorithmic thinking, multimodal and language-model systems, process mining, rigorous evaluation, and translating research into reliable agentic AI products.
Summary Generated by Built In
Be at the Forefront of the Agentic AI Revolution
At Skan AI, we are pioneering the context engine for human and agentic execution, bringing context from enterprise operators, systems, and processes to power how the world's largest organizations execute their most complex, mission-critical work.
Why Skan AI
We're in hyper-growth mode at exactly the right moment in history. As enterprises race to adopt agentic AI, we're uniquely positioned to deliver the clear signal they desperately need: a platform that trains and grounds AI Agents in trillions of real execution signals, enabling reliable, compliant automation of their most complex processes.
Backed by Dell Technologies Capital and other leading investors, we're the only company that can bridge the gap between AI's promise and enterprise reality, making us perfectly positioned to define the agentic era for modern enterprises.
Our diverse, collaborative team of 250+ innovators is solving category-defining challenges at the intersection of AI, process intelligence, and enterprise work. Diverse perspectives fuel breakthrough thinking, cross-functional collaboration is the norm, and our work directly transforms how Fortune 500 companies operate. We are shaping the future of work itself.
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 a Senior Applied Scientist on this team, you will provide technical leadership across teams and 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 own significant research and technical areas, taking heterogeneous data such as action logs, noisy screen traces, semi-structured documents, and variant-heavy workflows and building systems that recover canonical process structure, meaningful variations, decision rules, dependencies, exceptions, and provenance.
You will shape the architecture of these systems and set evaluation standards for representations consumed by downstream agentic execution systems. Those representations must preserve enough detail for an agent to act correctly while remaining structured, compact, transparent, and auditable. The same observational data and learned representations also contribute directly to our work on computer-use models for last-mile interaction with enterprise software. The problems here rarely arrive with clean formulations. You may begin with a branch the system consistently fails to recover, two superficially different execution paths that should collapse into the same process, or an algorithm that works across thousands of traces but fails at millions. Your job is to turn those failures into tractable research programs: identify the right abstraction, formulate hypotheses, design evaluations that isolate the problem, and develop methods that solve it. You will lead ambiguous investigations, make consequential architecture and evaluation decisions, and mentor other scientists and engineers through the work. Depending on the problem, this may draw from process and decision mining, graph algorithms, structured prediction, representation learning, information retrieval, program synthesis, model post-training and distillation, or agentic systems.
We care less about allegiance to a particular technique than about finding the simplest method that reliably recovers the underlying structure, scales, and succeeds in production.
Responsibilities
Process discovery and representation
- Lead the design and development of pipelines that discover, extract, and formalize end-to-end business processes from event logs, UI interaction traces, screen recordings, and unstructured documents
- Develop and guide algorithms for process variant detection, noise removal, and trace consolidation, turning noisy, high-cardinality action sequences into clean canonical representations
- Own core representation and architecture decisions so downstream agents receive the information they need while irrelevant execution-level variation is collapsed
- Study and establish principles for the tradeoff between abstraction and fidelity: determine when superficially different trajectories represent the same underlying behavior and when small differences encode materially different decisions
- Build retrieval and grounding systems that incorporate SOPs, documents, and domain knowledge when process understanding requires evidence beyond execution traces
- Set technical direction for major research areas, review experimental evidence, and mentor scientists and engineers working on related methods
Evaluation and reliability
- Own rigorous evaluation frameworks for process discovery and extraction, including LLM-as-Judge pipelines, conformance checking against ground-truth processes, and regression suites that catch subtle degradation
- Develop evaluation methods for settings where there may not be a single canonical ground truth, including behaviorally equivalent process representations
- Define quality bars and systematically test exception paths, conditional branches, loops, rare variants, and the failure modes that naive extraction approaches miss
- Make and communicate deliberate algorithmic and architectural tradeoffs so systems remain tractable across millions of traces and thousands of variants, including choices around data structures, approximation, indexing, and level of abstraction
- Diagnose failures in production, turn recurring failure patterns into new algorithmic or research directions, and drive improvements through validation and deployment
Collaboration
- Work closely with the computer-use agents team so the representations you produce are expressed at the right level of abstraction, contain the necessary execution context, and are robust enough for downstream autonomous execution
- Lead technical collaboration across research, engineering, and product to move promising methods from prototype to production and measure their effect on downstream agent behavior
- Mentor other scientists and engineers through problem formulation, experimental design, technical reviews, and productionization without serving as their people manager
- Represent the team's work through high-quality technical documents, external publications where appropriate, and clear communication with technical and non-technical stakeholders
Requirements
- Ambitious and Hungry. “Far and away the best prize that life offers is the chance to work hard at work worth doing.” - Theodore Roosevelt
- We screen for comparable experience and strong fundamentals, not exact tool matches. Equivalent technologies are welcome.
- Problem Decomposition and Algorithmic Thinking. You can take a loosely defined problem, such as recovering a reliable process from messy action logs, decompose it into tractable sub-problems, reason carefully about edge cases, and arrive at an approach. You are comfortable making and defending architecture decisions grounded in first principles about data structures, graph algorithms, approximation, indexing, and time and space complexity when algorithms must operate over millions of executions.
- ML / DL Breadth. Deep working knowledge across representation learning, sequence modeling, clustering, structured prediction, information retrieval, generative modeling, and modern foundation-model adaptation. You have the judgment to recognize when a problem calls for a learned model, an algorithmic solution, an LLM-based system, or some combination of the three, and can guide others through those choices.
- Cross-Domain Research Translation. You read research from adjacent areas including frontier AI, computer vision, language modelling, process mining, program synthesis, information retrieval, graph learning, and knowledge representation, and can judge whether it applies. You can identify the core idea inside a method, understand the assumptions that make it work, determine whether those assumptions hold in your setting, and adapt the idea into a robust production system.
- Research Judgment and Leadership. You can distinguish a difficult engineering problem from a research problem, define a coherent research direction, design experiments that isolate the important variable, raise the quality of others' work through mentorship and review, and abandon an elegant approach when the evidence suggests a simpler method works better.
- Applied Foundation Model Systems. Extensive hands-on experience building reliable systems with modern language or multimodal models, including retrieval, structured generation, evaluation, grounding, and model adaptation. You understand the failure modes that arise when model outputs are used to recover structured process knowledge from noisy evidence and have owned systems from research through production.
This may include experience with:
- Retrieval design, embedding strategies, chunking, re-ranking, and grounding against source evidence
- LLM-as-Judge systems, rubric design, calibration against human labels, and knowing when automated evaluation is insufficient
- Hands-on work with LLMs, RL, RLHF/RLAIF, post-training (SFT), evals, graders, synthetic data, model training, and production ML systems
- Faithfulness verification, constrained generation, confidence calibration, and methods for reducing unsupported model outputs
- Guardrails, input and output validation, graceful degradation, and monitoring for drift in production
- Python Proficiency. Excellent software engineering in Python, experience with ML frameworks such as PyTorch, comfort with data processing at scale, and a demonstrated ability to write clean, tested, production-quality code and establish sound engineering practices for research systems.
- Communication. You can explain an algorithmic or architectural tradeoff to both a researcher and a product manager, write clear technical and design documents, and provide precise feedback that improves the work of other scientists and engineers.
Preferred Experience
- A sustained research track record, demonstrated through influential publications in top-tier machine learning, artificial intelligence, reinforcement learning, natural language processing, computer vision, or data mining venues
- Experience leading research ideas from published work through production deployment with measurable impact
- Experience defining technical direction, mentoring researchers or engineers, and influencing architecture across teams
- 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 from unstructured text
- Multimodal data processing: extracting signals from screen recordings, UI screenshots, or interaction traces alongside textual data
- Foundation-model adaptation, including supervised fine-tuning, LoRA, synthetic-data methods, distillation, reinforcement-learning-based post-training, and evaluation of resulting model behavior
You Might Be a Great Fit If You
- Look for the hidden structure inside messy observations before deciding what to build. You are excited by the possibility of discovering how a process actually works without being handed its rules in advance.
- Enjoy problems where the right abstraction, model, and formulation are not obvious. You are willing to move between algorithms, learned representations, graph structures, retrieval, and foundation models until the system can explain what it has observed.
- Are drawn to the idea of building an end-to-end, process-agnostic machine: one that can observe unfamiliar work, infer its underlying boundaries, steps and decision logic, distinguish meaningful variation from noise, and produce concretized knowledge that gives an agent the full state representation to successfully navigate complex real-world processes.
- Read deeply across fields and adapt ideas from process mining, program synthesis, representation learning, graph algorithms, information retrieval, multimodal learning, and agentic systems. You care about the underlying insight more than the category it came from.
- Are comfortable moving between theory and reality: papers, hypotheses, experiments, raw traces, screen recordings, production failures, and systems that must work at organizational scale.
- Naturally take technical ownership, help others sharpen their thinking, and maintain a high bar for architecture, evaluation, publication-quality research, and production reliability.
- Want to work with a small, unusually ambitious team where research judgment and engineering execution are inseparable, and where your ideas can shape the architecture of an entirely new kind of system.
Skan AI is an equal opportunity employer committed to building a diverse, inclusive, and respectful workplace around the world. We do not discriminate based on race, color, religion or belief, sex (including pregnancy, sexual orientation, gender identity, or gender expression), national origin, ancestry, age, disability, medical condition, genetic information, marital or family status, military or veteran status, or any other characteristic protected by applicable laws in the locations where we operate.
We welcome people from all backgrounds and provide reasonable accommodations throughout the hiring process
Skills Required
- Strong problem decomposition and algorithmic thinking, including data structures, graph algorithms, approximation, indexing, and complexity analysis at large scale.
- Deep working knowledge of representation learning, sequence modeling, clustering, structured prediction, information retrieval, generative modeling, and foundation-model adaptation.
- Ability to evaluate and adapt research across frontier AI, computer vision, language modeling, process mining, program synthesis, information retrieval, graph learning, and knowledge representation.
- Research judgment and leadership, including research direction, experimental design, technical review, mentorship, and architecture decisions.
- Extensive hands-on experience building reliable language or multimodal model systems involving retrieval, structured generation, evaluation, grounding, and model adaptation.
- Excellent Python proficiency, experience with ML frameworks such as PyTorch, scalable data processing, and production-quality software engineering.
- Ability to communicate algorithmic and architectural tradeoffs to technical and non-technical stakeholders and write clear technical documentation.
- Sustained research track record demonstrated through influential publications in machine learning, artificial intelligence, reinforcement learning, NLP, computer vision, or data mining venues.
- Experience taking research ideas from publication through production deployment with measurable impact.
- Experience defining technical direction, mentoring researchers or engineers, and influencing architecture across teams.
- Experience with process mining, event-log analysis, process discovery, conformance checking, or variant analysis.
- Experience extracting structured information from documents, including NER, relation extraction, table extraction, or structured output generation.
- Experience processing multimodal data such as screen recordings, UI screenshots, interaction traces, and textual data.
- Experience with foundation-model adaptation, including SFT, LoRA, synthetic data, distillation, reinforcement-learning post-training, and model evaluation.
Am I A Good Fit?
Get Personalized Job Insights.
Our AI-powered fit analysis compares your resume with a job listing so you know if your skills & experience align.
Success! Refresh the page to see how your skills align with this role.
The Company
What We Do
Skan AI is an enterprise AI company that provides a process intelligence and automation platform. By observing how work actually happens, Skan creates a 'Context Graph of Work' and digital twins of operations. This allows Fortune 500 companies to understand the telemetry of their digital operations, enabling confident automation and lasting transformation through AI agents trained on operational reality.
