Section heading
Highlight 1
Highlight 2
Highlight 3
Heading
Value 1
Label 1
Value 2
Label 2
Value 3
Label 3
Section heading
Add the section description.
Section heading
Heading
Value 1
Label 1
Value 2
Label 2
Value 3
Label 3
Section heading
Section heading
Section heading
Heading
Value 1
Label 1
Value 2
Label 2
Value 3
Label 3
Highlight 1
Highlight 2
Highlight 3
Section heading
Section heading
Section heading
Add the section description.
Heading
Value 1
Label 1
Value 2
Label 2
Value 3
Label 3
Section heading
Heading
Value 1
Label 1
Value 2
Label 2
Value 3
Label 3
Making AI quality measurable at every stage
Braintrust set out to make AI quality measurable. We built the UI that turns thousands of eval results into something a team can compare and act on.
Built the web app for Braintrust’s early AI evaluation platform.
Created core workflows including the prompt playground and large-dataset visualizations.
Rapidly shipped new features and refinements based on early customer feedback.
AI outputs can change from one test to the next, making quality hard to measure. Braintrust needed to turn complex evaluation data into something teams could understand, compare, and act on.
A fast interface for a complex feedback loop
We built a clear interface for working with prompts, datasets, model responses, and evaluation results. A prompt playground made experimentation faster, while visualizations helped teams spot patterns across larger test sets.
Test, compare, improve
At the heart of Braintrust is a simple idea: improving AI should be an evidence-based process. Start with a dataset, define what good looks like, then test prompts and models against it.
Score the results, compare changes side by side, and use what you learn to improve the next version. Real-world failures can become new test cases, making each round of evaluation more useful.
Letting customer feedback shape the product
Braintrust was learning from customers while building the product, so priorities changed often. We worked inside Braintrust's team and owned features from spec to release. That gave a small team more capacity without adding a layer of handoffs or slowing down decisions.
“I’m a repeat client of Fluxon’s for a reason. The Fluxon team are experts in their field and deliberate in their work. They delivered a robust UI and helped us become a leading stack in development of AI systems.”
From early product to production AI infrastructure
Fluxon helped Braintrust get its early product to market quickly while its team stayed focused on the core technology and customers. Since then, Braintrust has grown into a platform used by leading AI teams at Notion, Dropbox, Coursera, Graphite and others, and in February 2026 the company announced an $80 million Series B to continue building infrastructure for production AI.
$80M Series B
raised in 2026
Thousands
of organizations use Braintrust
Leading AI teams
including Notion, Dropbox, and Coursera
.svg)
.avif)
