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
- Understanding the output of a decision model.
- Distinguishing system one decisions from reasoning LLMs.
- Analyzing the cost and latency profiles of bounded decisions.
The Open-Weight Decision Model Landscape
- AWS Labs’ Strands Decider 2B.
- JEV and GEV open-weight model families.
- Pplx-Decider, Liquid D1, and Clef.
- Comparing licensing and deployment variations.
Inside a Decision Model
- The LLM torso utilized without a language modelling head.
- The pointer head mechanism for scoring candidate options.
- Deriving confidence scores from the model.
Executing Your First Decisions
- Local setup on CPU and consumer-grade GPUs.
- Crafting questions and defining option sets.
- Processing multiple questions in a single pass.
- Configuring thresholds and fallback mechanisms.
Assessing Decision Quality
- JevBench and other decision-specific benchmarks.
- Balancing accuracy and calibration.
- Evaluating Brier score and confidence reliability.
- Identifying tasks where decision models are unsuitable.
Requirements
- Foundational knowledge of large language models and prompt-driven applications.
- Proficiency with command-line interfaces and Python scripting.
- No prior experience in machine learning training is necessary.
Target Audience
- AI engineers assessing decision models for agent workflows.
- Software developers seeking to substitute LLM calls with bounded decisions.
- Solution and platform architects designing agent architectures.
- Technical product managers evaluating decision model capabilities.
