Capital One Tech

I’m excited to announce that @CapitalOne is open-sourcing VulnHunter, an advanced agentic AI security tool we built to help the entire developer ecosystem find and fix software vulnerabilities. The digital threat landscape is moving faster than human defenders can keep pace. Because modern supply chains are deeply interconnected, a single vulnerability can ripple across thousands of enterprises simultaneously. No single organization can solve this challenge alone, and stakes are only rising for security teams to fight AI-enabled threats with equally capable AI-driven defenses to protect our digital environments. We designed VulnHunter with a developer-first mindset to solve a massive industry pain point: overwhelming false positives that create friction and slow down daily workflows. VulnHunter represents a new approach through three primary technical innovations:  -Falsification Engine: It actively tries to disprove its own findings to virtually eliminate false alarms. -Attacker-First Analysis: It simulates a hacker's journey, tracing entry points to map real attack paths. -Evidence-Backed Repair: It generates the exact code changes needed to fix verified defects. I’m incredibly proud of our internal team for exemplifying Capital One’s culture of innovation and invention. I’m also appreciative of our partnership with @AnthropicAI. VulnHunter’s advanced reasoning workflow is optimized to run using the exceptional capabilities of Claude Opus 4.8 and the Claude Code environment. It’s a noteworthy milestone in our collaboration, and I can’t wait to see how we’ll continue to create state-of-the-art capabilities together. At Capital One, we hope the release of VulnHunter enables the broader tech and security community to inspect the workflow, challenge its assumptions, and contribute improvements. Collectively using and contributing to these solutions can help advance the betterment of the systems we all build and the customers who depend on us to keep their data safe. Get the details here: i.capitalone.com/JyA3kUkDu

@AnthropicAI Dive into the full project and code on our GitHub repository: github.com/capitalone/vul…

Read more about the project on our tech blog. It is optimized for @AnthropicAI’s Claude Opus 4.8 and runs directly within Claude Code: i.capitalone.com/GWUe2ebrE

Traditional code scanners can flood developers with false positives. That’s why we built VulnHunter, an open-source code security tool with an agentic reasoning workflow. It leverages a unique falsification engine to challenge its own findings and simulate real attacker paths.

Check out our latest #AppliedAI research to see how we’re applying these insights to our proprietary #AI solutions. #AIResearch capitalone.science/blog/llm-reaso…

Live from #ICML2026! We’re thrilled to share that nine Capital One-contributed papers were accepted this year's conference!

Day 1 of #ICML2026 is underway in Seoul! We’re excited to engage with the machine learning community to discuss how we’re advancing #ML through proprietary AI solutions. If you're attending the conference, come visit us at booth B117. #MachineLearning #ICML

Our teams are continuously exploring new frontiers in machine learning and natural language processing to build proprietary AI solutions. Check them all out on our blog: capitalone.science/blog/capital-o…

We’re proud to have contributed to five research papers accepted at #ACL2026NLP this year, including one Capital One led paper being presented at the main conference! #NLProc #MachineLearning #NLP #AIResearch

We’re officially at #ACL2026NLP! Stop by booth D2 to meet our team and learn more about our proprietary AI solutions in machine learning. #NLProc #MachineLearning #NLP #AIResearch

Explore our latest peer-reviewed publications: i.capitalone.com/Guxn49l2f

We’re thrilled to launch Capital One Science, a new hub dedicated to our foundational #AI research. See how we're investing in partnerships and IP to advance state-of-the-art AI science.

A @forrester Consulting study commissioned by Capital One found that focusing on usability, adopting low/no-code interfaces wherever possible, and applying sensible governance policies are all ways to expand #machinelearning in the workplace. Learn more from the findings ⬇️

ML and data engineering 🤝 DevOps. Learn how MLOps can help streamline data processing and machine learning processes and foster collaboration across teams below.

Need a starting point for using the Pandas concat(), join() and merge() methods to perform data analysis? We’ve got you covered–read more below.

We wanted to make it easier for customers to redeem their Capital One credit card rewards in store. So we built a #serverless, scalable API that lets customers redeem their rewards at the point of sale in partner stores, like REI. Learn more here 👇

We built a digital pass system to help eligible customers gain contactless entry to Capital One Lounges. What looks like an easy customer experience is actually multiple #serverless, scalable APIs built in-house. See how we built it 👇

What do olympic-sized swimming pools and #angular have in common? Find out in software engineer Reagan Henke’s latest blog about her experience at @ngconf and what the latest Angular product changes and features mean for web development teams.

Reasons to avoid configuring SQS batch size of 1 for #Lambda with a SQS event sources: 1️⃣ It’s expensive 2️⃣ It can lead to higher processing times for high message volumes 3️⃣ It can exhaust available concurrency for Lambda functions in the AWS account. Learn more below ⤵️

Spread your wings and fly over to our latest blog about Flutter app development and techniques for developers building cross-platform apps.