Questions about getting started with SLAN.
About SLAN
SLAN turns expert knowledge and methods into personalised AI coaches that help people apply what they learn to real problems.
Nudge is SLAN's AI coach. It uses questions and feedback to help someone work through a problem using the guidance selected for that coach.
SLAN supports education, business and expert-led coaching. The experience is shaped around the intended audience and task.
The coaching conversation responds to the situation, information and answers a person shares. It doesn't imply a persistent learner profile or automatic long-term memory across sessions.
Nudge uses a chat interface, but it’s built to take learners through guided multi-stage flows: steps, decision checks, and completion criteria. The goal is application, not unlimited free-form Q&A.
No. It extends application support between sessions and across a defined audience. Plan where direct teaching, coaching and human review remain necessary.
Getting started
Examples include analysing a case, applying a business framework, planning a next step or practising a difficult conversation. The coach’s purpose depends on the methods and materials you choose.
Choose one use case, identify the relevant materials and agree who will use the coach. A demo helps us explore that setup together.
Start with relevant slides, PDFs, readings, notes or playbooks. Confirm file compatibility and permissions during setup, especially for licensed third-party content.
The expert helps define the purpose and reviews the proposed guidance. Allow time to test representative situations and refine the experience before sharing it.
Yes. Review the proposed frameworks and test the coaching guidance before making the coach available to your audience.
A typical pilot flow looks like:
- upload materials,
- extract candidate structures (frameworks and multi-stage flows),
- review and approve,
- publish to a limited cohort,
- iterate based on learner questions and outcomes.
Exact steps are often phased depending on pilot scope and governance requirements.
We'll discuss your audience, the task you want to support and the relevant materials, then explore whether coaching is a good fit.
Quality and control
It can be. AI-generated guidance can contain errors or miss important context. Review and test the coach, check important information against relevant sources, and give people a clear way to report problems.
The system is designed to reduce drift by:
- grounding answers in approved materials (where configured),
- keeping course intent explicit through authoring and review,
- scoping what content is visible per cohort and course,
- optionally showing what sources informed the answer (where feasible).
Exact techniques depend on deployment configuration and content permissions.
Yes. Review and refine its guidance as your methods develop and as you learn from the questions people ask. We agree together how updates are tested and released.
Coaching helps people develop conclusions and next steps in the conversation. Exportable artefacts are coming soon, so don't assume they're available in a pilot.
Access and data
Yes. SLAN supports access organised around courses, assistants and cohorts. Choose the joining and access settings that fit the deployment.
Students can join with a registration code or invite link. Enterprise plans add single sign-on and your school's branding; any LMS integration is scoped with you.
Follow the invitation or joining instructions from your instructor or programme team. Existing users can sign in to the coaches available to them.
SLAN’s application and data storage can be hosted in AWS eu-central-1 (Frankfurt). Model inference is provided by third-party model providers; processing location depends on the provider and endpoint, and may be outside the EU. If EU-only processing is a hard requirement, we treat it as a deployment requirement: EU-hosted model options or customer-managed patterns.
No. SLAN's published policy states that submitted content is not used for model training. We use Google Gemini, Anthropic (Claude) and OpenAI through their paid API terms, not free tiers.
Your materials stay yours. You choose who can use each coach, by course, cohort or organisation, and what they can see. You can keep your source files internal, so people get the coaching without the raw materials. You set these rules for each coach.
Review the proposed data, access, model processing and contractual requirements before sharing sensitive materials. The governance page has the current details.
Pilots and pricing
Combine available usage information with faculty review and student feedback. Agree the measures and data collection before launch; don't treat activity alone as evidence of learning.
It depends on how many people use the coaches and how you run them, for example per person, per cohort or by usage. We'll talk through your setup and quote for it.
For your setting
The permitted role of AI should follow your assessment rules. Agree what the coach may support and test it against representative student requests before release.
You can start with materials you already use, then review how they are translated into coaching guidance. Only share content you are permitted to use for this purpose.
Training is one starting point. A coach can also be scoped around applying an expert method during a work task or while exploring a business problem.
No. The coaching supports people's analysis and next steps. Your organisation remains responsible for decisions and any required human review.
No. It's an AI coach built around the knowledge and guidance you choose to share. It is presented clearly as an AI experience, with a defined purpose and boundaries.
Access can be organised around a defined cohort. Discuss your delivery and commercial model in the demo; payment collection, subscriptions and revenue sharing need separate confirmation.
Still have questions?
Tell us who you want to support and what you want them to be able to do. We'll answer for your setting.