World Congress 2025 Aug 20, 2025 Session details

Bringing AI Model Testing and Prompt Management to Your Codebase with GitHub Models

Kevin Lewis , Sandra Ahlgrimm

Stop relying on vibes for prompt engineering. Treat AI evaluation with strict determinism. GitHub Models brings CI/CD and version-controlled testing directly into your codebase.

Pause
Mute Enter Fullscreen
#1 about 3 min

The challenge of testing AI prompts without tools

Why developers rely on vibe checks instead of systematic tests for non-deterministic AI outputs.

#2 about 1 min

Introduction to the GitHub Models testing toolkit

An overview of accessing over 40 models using a single API key for prompt experimentation.

#3 about 2 min

Managing and versioning system prompts as YAML files

How to configure system instructions and parameters using YAML configuration directly in your repository.

#4 about 3 min

Configuring generation parameters and committing prompt changes

Adjusting temperature and token limits before committing prompt configurations to feature branches.

#5 about 3 min

Evaluating prompt accuracy against expected test outcomes

Establishing test datasets and using LLM judges to ensure accuracy and coherence.

#6 about 3 min

Running simultaneous comparisons across different AI models

Experimenting with multiple models and prompt variants simultaneously to optimize latency and cost.

#7 about 2 min

Integrating GitHub Models into external enterprise applications

Consuming managed prompts using model API endpoints or the Azure AI Inference SDK.

#8 about 4 min

Building applications with Azure AI Inference SDK

Code walkthrough showing how a JavaScript application imports an SDK to consume repository configurations.

#9 about 2 min

Evaluating AI models locally using GitHub CLI

Running interactive sessions and local evaluations directly from the terminal before committing code changes.

#10 about 4 min

Automating repository issues and changelogs with AI actions

Leveraging inference integrations inside GitHub Actions to automate pull request summaries and triage bug reports.

#11 about 1 min

Running prompt evaluation tests in CI/CD pipelines

Running automated evaluations during pull requests prevents degraded prompts from reaching production pipelines.

#12 about 3 min

Migrating existing applications to managed prompt architectures

Converting plain text instructions to standard YAML files dramatically accelerates testing and deployment workflows.

Matching moments

1:55 min

Evaluating AI prompts as code with automated CI pipelines

Akshay Nagpal Akshay Nagpal · Europe 2026 Virtual

4:04 min

Introduction to GitHub Next and AI prototyping

Krzystof Czieslak · LIVE

1:11 min

Testing AI models quickly using GitHub Models

Ricardo Ricardo · World Congress 2025

1:48 min

Exploring AI integrations in modern agile development workflows

Markus Walker Markus Walker · World Congress 2023

1:31 min

The shift from human prompt engineering to AI-generated prompts

Vitaly Friedman Vitaly Friedman · World Congress 2026 Europe

2:48 min

Integrating version control and AI into productivity software

Upcoming sessions on this topic

Open session

World Congress 2026 North America

September 24, 2026 · 10:20–10:50

Stage 6

Practices, Not Prompts: Scale GitHub Copilot with AI-Ready Repositories

Luis Pujols

Staff Customer Success Architect, GitHub

Luis Pujols
Open session

World Congress 2026 North America

September 24, 2026 · 16:10–16:40

Stage 1

Sandboxing the Swarm: Building Secure, Serverless AI Agents with Wasm

Lena Hall, Thorsten Hans

Lena Hall
Thorsten Hans
Open session

World Congress 2026 North America

September 23, 2026 · 15:40–16:10

Stage 2

Lean Intelligence: Lessons from GitHub Copilot Data Science Efforts

Rahul Pandita

Researcher and Technical Advisor at Microsoft

Rahul Pandita
Open session

World Congress 2026 North America

September 23, 2026 · 13:40–14:10

Stage 2

Building an AI-Native Development Workflow

Vanessa Minik

Senior Service Delivery Engineer, GitHub

Vanessa Minik
Open session

World Congress 2026 North America

September 24, 2026 · 14:25–14:35

Outdoor Stage

SecurePrompt: Building a Pre-Flight Security Layer for Agentic AI

Ravi Sastry Kadali

AI/ML Engineer at General Motors

Ravi Sastry Kadali
Open session

World Congress 2026 North America

September 24, 2026 · 11:00–11:30

Stage 6

Accelerating development and managing agents with the GitHub Copilot app

Christopher Harrison

Senior Developer Advocate at GitHub

Christopher Harrison