15 years of experience in the networking field, specializing in Software-Defined Networking (SDN), I have recently delved into the world of AI/ML. My focus has been on researching the applications of Large Language Models (LLMs) in networking, exploring how these advanced technologies can supercharge networking and user experiences.
Experiences
Most recent first
- 2 years working in a startup building a Networking AI/ML product
- 4 years as solutions architect / consultant / pre-sale in networking, specifically Cisco ACI and DataCenter
- 5 years as a network engineer in Enterprise covering a wide range of technologies - DC/Campus Networking, CDN, Automation
- 5 years as a tier-1 tech support engineer in an ISP setting
White Papers Published
- Segmenting Complex Environments Using Cisco ACI
- Demystifying ACI Application Centric “Mode” Through the Lens of an Engineer
- Cisco ACI: Design to Automate
Career Certifications and Awards
- Patent: US11606265B2 - Network control in artificial intelligence-defined networking
- Awarded by WWT as Engineer of the Month/Quarter/Year in 2021
- Cisco Certified Network Associate
- Cisco Certified DevNet Associate
- Cisco Certified DevNet Professional
- Cisco Certified DevNet Specialist - Core
- Cisco Certified DevNet Specialist - Data Center Automation and Programmability
- Coursera - Generative AI with LLM
- Coursera - Generative AI for everyone
- Coursera - Machine Learning
- Coursera - AI For Everyone
My Thoughts On AI
My primary goal is to harness the power of LLMs to facilitate the integration of AI/ML into data center designs, effectively bridging the gap between high-level business requirements and network functionalities. This approach is rooted in the philosophy of true intent-based networking, which prioritizes understanding and fulfilling business and application needs, rather than focusing solely on the network infrastructure.
I believe AI is here to augment, not to replace network engineers. It’s akin to the calculator’s role in mathematics: a powerful tool that expands capabilities without diminishing the importance of human expertise. The calculator didn’t replace mathematicians; instead, it made math more accessible and accelerated our learning.
Similarly, while the core principles of networking remain crucial, our approach evolves. Engineers need a solid grasp of these fundamentals to discern when an LLM might ‘hallucinate’. The shift we’re witnessing moves us away from rote memorization of CLI commands towards a focus on design and business alignment. Our goal is to provide nuanced, business-specific directives to LLMs, enhancing efficiency and customization.
Envision a future where networking operations are driven more by human language interfaces than traditional CLI/GUI. This paradigm shift promises a more intuitive, human-centric approach to networking technology
– written by me, revised by chatGPT ( of course )