Unlocking Networking AI Innovation: Introducing Essedum
summary
Essedum is an open-source Essedum framework that stitches MLOps, cloud connectors and accelerator APIs via REST into deployable network AI apps, donated to LF Networking and published on GitHub; it runs pipelines, handles lifecycle and supports continuous production deployment. A separate Salis responsible AI toolkit provides security, privacy and fairness checks, and a data anonymization plugin ingests telco datasets without exposing PII. Example integrations include SoftBank’s AITR packaged with NVIDIA and show end-to-end deployment flow rather than just prototypes.
AI-written summary · about
chapters (11)
- 0:00Domain-specific AI in networking
- 1:05Telecom infrastructure and AI needs
- 2:32Linux Foundation's AI framework
- 5:02Productizing domain-specific AI
- 6:46SODM simplifies AI development
- 8:14MLOps and data pipelines in SODM
- 9:18Data anonymization for privacy
- 10:21Introducing Salis for responsible AI
- 12:10SoftBank's AITR real-world example
- 13:38Customization and open-source integration
- 14:02Summary and next steps
related talks
38mIntroducing Open Platform for Enterprise AI
12mUsing AI-enabled Edge Cloud Technologies to Strengthen Europe’s...
05mEdge AI: Because the Cloud is Too Slow...
29mOpen Source Sovereign AI: Building Enterprise GenAI Platforms for Control & Comp...
11mKeynote: LF Edge: Open Source to Enable 5G, IoT and AI Development
38mIntroducing Open Platform for Enterprise AI
37mEnabling AI For Everyone - Hiai Open Platform Introduction
16mAI usage in telcos
