Datacurve AI
Datacurve AI builds tools and benchmarks for evaluating and improving frontier coding agents. The team develops DeepSWE, a contamination-free long-horizon software engineering benchmark, and Pier, a sandboxed coding-agent evaluation framework. Their work focuses on rigorous, real-world measurement of AI coding capabilities across multiple programming languages and open-source repositories.
At a Glance
Discussions
No discussions yet
Be the first to start a discussion about Datacurve AI
Know more about Datacurve AI? Start a discussion to share what you know.