Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Duration 7 hours
Course Outline
From Text Generation to Typed Decisions
- The nature of outputs returned by decision models
- System one decisions compared to reasoning LLMs
- The cost and latency profile of bounded decisions
The Open-Weight Decision Model Landscape
- Strands Decider 2B from AWS Labs
- JEV and GEV open-weight families
- Pplx-Decider, Liquid D1 and Clef
- Differences in licensing and deployment
Inside a Decision Model
- LLM torso without a language modelling head
- Pointer head scoring over candidate options
- The origin of confidence scores
Running Your First Decisions
- Local setup on CPU and consumer GPUs
- Structuring questions and option sets
- Processing multiple questions in a single pass
- Managing thresholds and fallback behaviour
Evaluating Decision Quality
- JevBench and other decision benchmarks
- The trade-off between accuracy and calibration
- Brier score and the reliability of confidence
- Scenarios where a decision model is not the appropriate tool
Requirements
- Basic familiarity with large language models and prompt-based applications.
- Confidence in using the command line and writing Python scripts.
- No prior experience in machine learning training is required.
Audience
- AI engineers assessing decision models for agent workflows
- Software developers seeking to replace LLM calls with bounded decisions
- Solution and platform architects designing agent architectures
- Technical product managers evaluating decision model capabilities
