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
Transitioning from Text Generation to Typed Decisions
- Understanding the output structure of decision models.
- Distinguishing system one decisions from reasoning LLMs.
- Analyzing the cost and latency characteristics of bounded decisions.
The Landscape of Open-Weight Decision Models
- AWS Labs' Strands Decider 2B.
- The JEV and GEV open-weight model families.
- Alternatives including Pplx-Decider, Liquid D1, and Clef.
- Differences in licensing and deployment strategies.
An Inside Look at Decision Model Architecture
- The LLM torso structure without a language modelling head.
- Pointer head mechanics for scoring candidate options.
- Origins of the confidence score calculation.
Implementing Your First Decisions
- Local configuration on CPU and consumer-grade GPUs.
- Formulating effective questions and option sets.
- Processing multiple questions in a single pass.
- Managing thresholds and fallback behaviors.
Evaluating Decision Quality
- Reviewing JevBench and other decision benchmarks.
- Balancing accuracy against calibration.
- Interpreting Brier scores and confidence reliability.
- Recognizing tasks where a decision model is unsuitable.
Requirements
- Fundamental understanding of large language models and prompt-based applications.
- Proficiency with the command line and Python scripting.
- No prior experience in machine learning training is necessary.
Target Audience
- AI engineers assessing decision models for agent workflows.
- Software developers substituting LLM calls with bounded decision processes.
- Solution and platform architects designing agent-based architectures.
- Technical product managers evaluating options for decision models.
