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.
- Contrasting System One decisions with reasoning LLMs.
- Analyzing the cost and latency advantages of bounded decision tasks.
The Current Landscape of Open-Weight Decision Models
- Overview of Strands Decider 2B from AWS Labs.
- Introduction to the JEV and GEV open-weight model families.
- Review of Pplx-Decider, Liquid D1, and Clef models.
- Comparing licensing requirements and deployment variations.
Architectural Deep Dive: Inside a Decision Model
- Utilizing an LLM torso devoid of a standard language modeling head.
- Implementing a pointer head for scoring candidate options.
- Tracing the origin and calculation of confidence scores.
Practical Implementation: Running Your First Decisions
- Local deployment setup on CPUs and consumer-grade GPUs.
- Best practices for phrasing questions and defining option sets.
- Processing multiple questions in a single inference pass.
- Configuring thresholds and managing fallback behaviors.
Evaluating Decision Quality and Performance
- Utilizing JevBench and other relevant decision benchmarks.
- Balancing accuracy against calibration metrics.
- Applying the Brier score to assess confidence reliability.
- Recognizing scenarios where a decision model is not the appropriate tool.
Requirements
- Foundational knowledge of large language models and prompt engineering 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 integration into agent workflows.
- Software developers looking to replace broad LLM calls with bounded, specific decisions.
- Solution and platform architects designing robust agent architectures.
- Technical product managers evaluating options for decision model adoption.
