Scorecard AI
Artificial Intelligence · AI Agents
Overview
Scorecard AI builds evaluation infrastructure for teams shipping products on top of large language models. It runs test sets, scores model output against defined criteria and tracks regressions between releases. The product aims to give AI features the same release discipline as ordinary software.
Visa sponsorship
1
H-1B filings in 12 months
$218k
Median offered salary
0
Open roles in United States
Who they sponsor
Software developers
1
In the register as Scorecard AI t/a Scorecard, San Francisco, CA
Other sponsors in the same field
Fastino 
Lapis 
RapidFire AI 
Vidoso
Alert me
US Department of Labor · latest filing Jun 23, 2026 · Visa details by country
News
Report
Making research papers feel brainrot so I can read them Scorecard Blog
I built a bot that turns research papers into brainrot and it sucked. Here's how I fixed it.
Read more
Report
Annotation is all you need Scorecard Blog
Learn how structured, span-level annotations on OpenTelemetry traces create a continuous improvement loop for AI agents - turning human feedback into targeted code fixes via MCP-connected coding assistants.
Read more
Report
The First Zero-Code Tracing Setup for the Claude Agent SDK Scorecard Blog
Reveal Anthropic's hidden reasoning steps in 0 lines of code.
Read more
Report
You can't QA your way to the frontier. Scorecard Blog
QA finds what's broken. It doesn't help you get better. The best AI teams are shifting from "observe and evaluate" to "simulate and improve." Here's what Waymo, OpenAI, and frontier labs figured out about building self-improving agents.
Read more
Report
Introducing AgentEval.org Scorecard Blog
Introducing AgentEval.org: An Open-Source Benchmarking Initiative for AI Agent Evaluation
Read more
Report
Offices
Where the company hires and what each office is for
Cities
1
Hiring now
0
Countries
1
San Francisco
United States
HQ
10-50 est. staff
995 Market Street, San Francisco, CA 94103