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.
Course Outline
LLM Application Architecture and Design
- Standard OpenAI application patterns utilized for assisting personnel, augmenting capabilities, and automating workflows
- Selecting architectural frameworks that align with operational requirements, system reliability, and service user experience
- Transitioning from prototype development to scalable, maintainable application design for government use
Prompt Engineering, Context Management, and Structured Outputs
- Structuring system, user, and developer instructions to ensure predictable and consistent model behavior
- Designing prompts that enhance task control, consistency, and clarity in model responses
- Implementing structured outputs to facilitate downstream application logic and data integration
- Managing context windows, conversation state, and response quality to optimize performance
Tool Integration and Workflow Orchestration
- Utilizing function calling and tool-enabled workflows to interact with external services and systems
- Validating inputs and outputs, managing error states, and implementing fallback mechanisms
- Designing multi-step workflows to address complex business and operational tasks
Data Retrieval and Knowledge Grounding
- Identifying appropriate use cases for retrieval-augmented generation (RAG) strategies
- Preparing and chunking documentation to enable effective and relevant information retrieval
- Retrieving pertinent context to ground model responses in verified and trusted sources for government operations
Evaluation, Safety Guardrails, and Operational Readiness
- Establishing quality criteria and validating workflows against expected operational outcomes
- Mitigating hallucinations and addressing unsafe, irrelevant, or ambiguous queries through established guardrails
- Monitoring key performance indicators, including usage metrics, latency, token consumption, and cost efficiency
- Preparing applications for secure deployment, ongoing support, and iterative improvement
Practical Implementation Workshop
- Developing a comprehensive end-to-end OpenAI application that integrates prompting, structured outputs, tool use, and data retrieval
- Evaluating design decisions, addressing common implementation challenges, and defining practical next steps for production deployment
Requirements
* Demonstrated understanding of large language model (LLM) architectures and the development of applications utilizing application programming interfaces (APIs).
* Proficiency in employing REST APIs, JavaScript Object Notation (JSON), and workflows driven by prompt engineering.
* Intermediate proficiency in programming languages such as Python or JavaScript.
**Target Audience**
* Software engineers tasked with the development of applications leveraging LLM capabilities.
* Artificial intelligence engineers and technical leads responsible for designing solutions integrated with OpenAI technologies.
* Product management teams and solution architects accountable for the deployment and maintenance of production-grade artificial intelligence features for government.
7 Hours
