The judgment to know what to build, and the engineering to ship it.

Applied AI is Stefan Jansen's consulting practice for regulated and data-intensive organizations. For over a decade, he has helped teams move from AI ideas and pilots to governed production workflows.

Stefan Jansen, principal

Stefan Jansen

Stefan works on the strategy, architecture, evaluation, and code. He came to machine learning through quantitative economics, financial markets, and the business side of technology, then added the engineering depth to build the systems himself. That sequence shapes the practice: first establish which decision or workflow should improve, then choose and build the approach that can be evaluated, governed, and operated.

His work spans agent workflows, contract intelligence, forecasting, customer-lifecycle systems, and quantitative research infrastructure. He is also the author of Machine Learning for Trading, now in its third edition, and has taught more than 100,000 professionals and learners. He holds a Harvard master's in economics and public policy, a Georgia Tech MS in computer science, and the CFA charter.

Selected work

Work across insurance, healthcare, and financial markets, including production systems and the advisory and teaching around them. Clients are named where the work is public or cleared; otherwise they are described by domain.

  • Contract intelligence for Swiss Re's digital insurance business ↗

    Built core components of a system that restored the structure of complex legacy policy documents, extracted detailed attributes, and turned them into transparent, auditable eligibility logic. It used transformer language models with domain adaptation in 2020, reached production, cleared internal governance review, and is covered by a Swiss Re patent application.

  • Forecasting and customer-lifecycle systems for RxSense / SingleCare

    Forecasting, retention, customer-lifecycle, and campaign-analytics systems for pharmacy-technology and healthcare workflows.

  • Data science across Loeb Enterprises and its portfolio companies

    Advised portfolio-company teams on machine learning and statistical modeling, with a focus on healthcare and consumer analytics: customer targeting, campaign analysis, model design, and building out in-house data-science capability.

  • Live trading infrastructure

    Designed, audited, and operated research and execution infrastructure across equities and digital assets, including live systems for a regulated European investment firm.

  • Executive AI education for a Fortune 100 insurer

    Designed and delivered executive education on AI, data science, and analytics for senior leadership.

  • Agent and forecasting systems

    Agent workflows designed around explicit evaluation, governance, and operating requirements rather than treating the agent as the unit of value.

  • Machine Learning for Trading ↗

    Author of the ML4T book, now in its third edition, and its open-source ecosystem.

How we engage

Discovery

1–2 weeks

We map the workflow, find the real constraint, and size the opportunity. You get a prioritized plan with a clear read on feasibility and value, without a long assessment.

Embedded delivery

3–12 months

We join your team to design, build, and ship the system, owning delivery while transferring knowledge. Most engagements put the first production slice in front of users within weeks.

Advisory

Ongoing

Architecture reviews, technical direction, and vendor evaluation for teams building AI capability in-house.

Schedule a call