The Collective Knowledge technology (CK) was developed by Grigori Fursin to help researchers and practitioners collaboratively benchmark, optimize and co-design complex computational systems (AI, ML, quantum, IoT) in terms of speed, accuracy, energy and different costs. CK also supports Grigori's vision to enable self-optimizing, brain-inspired and efficient computer systems with the help of an open repository of knowledge, a common research SDK with reusable best practices and DevOps, live SOTA scoreboards and reproducible papers ^{2009, 2014, 2017, 2019, 2020, overview paper.} . We would like to thank all our academic and industrial partners for their feedback and support!

Even though the CK technology is used in production for more than 5 years, it is still a proof-of-concept prototype requiring further improvements and standardization. Depending on the available resources, we plan to develop a new, backward-compatible and more user-friendly version - don't hesitate to get in touch if you are interested to know more!

Get in touch with the CK founder and lead developer

• Grigori.Fursin@cTuning.og (non-profit cTuning foundation)
• gfursin@cknowledge.io (cKnowledge SAS)

Get in touch with the CK community

• Join the CK Slack channel.


• Follow our news on Twitter:
  • Grigori Fursin
  • Automatic announcements about new components and results


• Join technical and research discussions at the Collective Knowledge Google group.


• Join the Artifact Evaluation google group to participate in discussions related to reproducible research at systems and ML conferences.


• Join our LinkedIn group on collaborative and reproducible deep tech research.