Coffee With Developers • Aug 17, 2026

Validating Real-World Usefulness of AI Models - Julia Kasper

Julia Kasper

There is no universally superior AI model. Frontend developers prefer Claude, while backend engineers favor GPT. Julia Kasper explains how to rigorously evaluate AI for your IDE.

Pause
Mute Enter Fullscreen
#1 about 2 min

Introduction to AI features in Visual Studio Code

An overview of launching AI models and Copilot capabilities inside the code editor.

#2 about 2 min

Developer adoption of integrated AI chat features

How traditional developers and AI-first coders utilize new integrated chat interfaces differently.

#3 about 3 min

Integrating new third-party AI models into environments

How the team manages quick turnarounds and policy restrictions for incoming AI model releases.

#4 about 3 min

Evaluating non-deterministic AI models with offline testing

Using traditional assertion-based test scripts to measure the reliability of non-deterministic AI responses.

#5 about 3 min

Choosing AI models based on task and personality

Balancing token costs and task efficiencies by evaluating benchmarks and specific model characteristics.

#6 about 2 min

Tracking model quality improvements and potential saturation

Observing incremental performance gains across updates despite a sense of overall AI capability saturation.

#7 about 2 min

Connecting custom local models versus optimized built-in providers

How developers can bring their own API keys while native models benefit from specialized system prompt optimization.

#8 about 3 min

Preparing system prompts and evaluating models before launch

Refining prompts and running internal dogfooding sessions to validate model quality before public release.

#9 about 1 min

Performing local A/B testing across multiple AI agents

Running parallel sessions with different local and remote agents to directly compare model outputs.

#10 about 2 min

Inspecting open-source system prompts to understand model behavior

Reviewing public prompts allows external contributors to suggest tweaks and better understand AI nuance.

#11 about 2 min

Navigating user questions on selecting the optimal model

Addressing developer inquiries by recommending the latest updates and switching models based on task context.

#12 about 2 min

Maintaining an open-source core amidst proprietary editor forks

Navigating the challenges of third-party forks while continuing to value the open-source community mindset.

#13 about 3 min

Leveraging AI to lower the barrier for external contributions

Using AI tools to explain large codebases empowers external developers to make their first pull requests.

Matching moments

1:48 min

Exploring AI integrations in modern agile development workflows

Markus Walker Markus Walker · World Congress 2023

2:41 min

Leveraging AI tooling for accelerated learning and productivity

Dan Mindru · Coffee With Developers

1:56 min

Analyzing cloud-based AI code completion architectures

Daniel Savenkov Daniel Savenkov · World Congress 2024

5:16 min

Motivations for adopting AI to enhance developer productivity

1:55 min

Shifting developer workloads and realistic AI productivity gains

Chris Heilmann +2 · LIVE

2:26 min

Customizing developer tools and integrating domain-specific AI agents

Francisco Carneiro Francisco Carneiro · World Congress 2025

Upcoming sessions on this topic

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 · 16:10–16:40

Stage 6

The Reality of AI Adoption in Enterprises

Priyanka Vergadia

Cloud and AI expert @Microsoft, previously lead Devrel @Google

Priyanka Vergadia
Open session

World Congress 2026 North America

September 24, 2026 · 15:30–16:00

Stage 7

Beyond Vibe Coding: Using CLI Tools as Your AI Counterpart in Pair Programming

Ron Veen

Java enthusiast and Special agent for Team Rockstars IT

Ron Veen
Open session

World Congress 2026 North America

September 24, 2026 · 16:50–17:20

Mainstage

Building AI Products vs. Building With AI

Aparna Dhinakaran, Rukmini Reddy, Tamar Bercovici

Aparna Dhinakaran
Rukmini Reddy
Tamar Bercovici
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 25, 2026 · 15:30–16:00

Stage 5

21 Experiments in Six Weeks: A Playbook for Improving Your AI Agent

Sofia Rest

Software Engineer at Sentry

Sofia Rest