Feature Management
Experimentation
Data Warehouse
Dynamic Configs
Metrics
Holdouts
Insights
Dashboards
Autotune
Documentation
Walkthrough Guides
SDKs and APIs
Open Source Code
Integration
Product Updates
How Statsig Works
Build vs. Buy
Experimentation Community
Peak Velocity
Customer Stories
A/B Testing Calculator
Feature Flags Liberated
Optimizing AI Model Temp
What is Product Observability?
Experiments with Generative AI
Experimentation Platforms
CUPED Explained
Feature Management
Experimentation
Data Warehouse
Dynamic Configs
Metrics
Holdouts
Insights
Dashboards
Autotune
Documentation
Walkthrough Guides
SDKs and APIs
Open Source Code
Integration
Product Updates
How Statsig Works
Build vs. Buy
Experimentation Community
Peak Velocity
Customer Stories
A/B Testing Calculator
Feature Flags Liberated
Optimizing AI Model Temp
What is Product Observability?
Experiments with Generative AI
Experimentation Platforms
CUPED Explained
AI Experimentation: How to optimize the performance of AI-powered features using Statsig (Webinar Recording)
Morgan Scalzo
Community and Event ManagerThis month we tapped Statsig’s resident AI experts, Joe Zeng and Skye Scofield, to discuss the aforementioned hottest topic in tech.
Over the last couple of months, our customer conversations on AI experimentation have increased dramatically (as you may have guessed), so we decided the time is now to dive deep into how experimentation can only benefit the development of AI features.
Enjoy this on-demand viewing and we hope you can join us live in the future!
AI Experimentation: How to optimize the performance of AI-powered features using Statsig
About this virtual meetup:
Got AI on the brain? Statsig can help you run experiments with AI apps, using our recent launch of Statbot. During this Learning Lab we'll teach you how to record important model inputs and outputs, such as prompt, model choices, cost, and latency.
Additionally, we'll provide some tips on how to measure your application's performance and interpret the results to make data-driven decisions.
Ready to try Statsig for free?
Sign Up
Recent Posts
Why people love Statsig’s customer support
Thanks to our support team, our customers can feel like Statsig is a part of their org and not just a software vendor. We want our customers to know that we're here for them.
What migrating experimentation platforms actually entails
Migrating experimentation platforms is a chance to cleanse tech debt, streamline workflows, define ownership, promote democratization of testing, educate teams, and more.
Understanding sample size
Calculating the right sample size means balancing the level of precision desired, the anticipated effect size, the statistical power of the experiment, and more.
Recency bias: The silent saboteur in statistics
The term 'recency bias' has been all over the statistics and data analysis world, stealthily skewing our interpretation of patterns and trends.
Early startup journey: Year two at Statsig
A lot has changed in the past year. New hires, new products, and a new office (or two!) GB Lee tells the tale alongside pictures and illustrations:
CUPED Explained
A deep dive into CUPED: Why it was invented, how it works, and how to use CUPED to run experiments faster and with less bias.