The Art of Network Engineering

Send us a textRecorded on-site in Austin, Texas, at AutoCon 4 (Network Automation Forum), Andy sits down with Colin Doyle to talk about the human side of technical communication and why it matters more than ever in technical careers.They dig into practical speaking advice for engineers: how to slow down without losing authority, why “dead air” feels scarier than it is, how to stop relying on scripts, and how to structure a talk so your audience can repeat your message when you leave the room. Colin shares the “audience-first” mindset shift: don’t tell your story, tell the audience’s story with you in it.Then the conversation widens into the network automation adoption problem: why network automation still lags behind other IT domains, why tooling fragmentation creates anxiety (“what if I learn the wrong thing?”), and why starting with Python is often the safest first step. Colin also reframes overlays (EVPN/VXLAN) as a fundamental shift: abstraction changes operations, pushes configuration to the edge, and makes intent-based operations and assurance the real job.If you’re a CLI lifer preparing to level up, or you’re giving your first big talk, this episode is a practical, grounding guide.In this episode: communication fundamentals, talk prep, booth culture at AutoCon, automation adoption barriers, overlays → intent → assurance, and why you don’t need to be a “kung fu wizard” to start automating.This episode has been sponsored by Meter. Go to meter.com/aone to book a demo now! You can support the show at the link below.Support the showFind everything AONE right here: https://linktr.ee/artofneteng

Send us a textMost network engineers know BGP, OSPF, and maybe EIGRP, but far fewer have hands-on experience with ISIS. In this episode of The Art of Network Engineering, Andy Lapteff sits down with Russ White and Mike Bushong for a deep, opinionated, and refreshingly honest discussion about routing protocol design in modern data centers.We explore why BGP has become the default hammer for every networking nail, what we lose when we blend underlay and overlay into a single protocol, and why some of the largest networks in the world still rely on IS-IS for simplicity, scale, and resilience.This isn’t a “which protocol is best” argument, it’s a design conversation. One about failure domains, operational reality, education gaps, and why many engineers never learn the protocols that quietly power hyperscale networks.In this episode:Why BGP is policy-rich but intentionally slowThe architectural value of separating underlay and overlayHow ISIS works and why it’s simpler than you thinkTLVs, scalability, and protocol evolutionWhy familiarity often beats good design (for better or worse)Where RIFT fits and where it doesn’tThe cost of losing deep protocol knowledge as engineers retireIf you’ve ever wondered why networks are designed the way they are, or if you’ve felt uneasy about “just using BGP everywhere,” this conversation is for you.Subscribe for more conversations where technology meets the human side of IT.This episode has been sponsored by Meter. Go to meter.com/aone to book a demo now! You can support the show at the link below.Support the showFind everything AONE right here: https://linktr.ee/artofneteng

Send us a textIn this special on-location episode of The Art of Network Engineering, Andy Lapteff sits down in person with Kevin Myers for a conversation that pulls no punches.Kevin brings decades of service provider and whitebox experience to the table as the two dive deep into one of networking’s most complex decisions: how to choose the right vendor to build your network.From Cisco to whitebox, from enterprise carpeted IT to hyperscale data centers, this episode is all about designing networks that align with business needs, not just personal bias or legacy choices.Topics include:Why vendor selection should come after requirements gathering.How multi-vendor environments can create both resilience and complexity.When whitebox networking makes sense—and what kind of teams can support it.The hidden "operational tax" of expanding your vendor portfolio.Why understanding the business is the most critical skill for modern engineers.Whether you're a design engineer, network architect, or just trying to future-proof your ops, this episode is packed with insights that will change how you think about gear, vendors, and the networks we build.This episode has been sponsored by Meter. Go to meter.com/aone to book a demo now! You can support the show at the link below.Support the showFind everything AONE right here: https://linktr.ee/artofneteng

Send us a textAndy sits down with longtime friend William Collins to unpack three big themes shaping modern NetOps. First: the AWS US-East-1 outage and the myth that “cloud = resiliency by default.” They explore blast radius, hidden regional dependencies, cost trade-offs (active/active vs. DR), and why resiliency is engineered, not purchased. Next: how public speaking accelerates a technical career (without live-demo heartbreak). William shares practical tactics to craft a memorable talk, lean on story, and handle Q&A. Finally: a plain-English walkthrough of Model Context Protocol (MCP), why it exists, how it standardizes tool access for LLMs, and what that means for real NetOps workflows. If you design for failure, want to level up your communication skills, or keep hearing “MCP” and wonder what it actually does, this one’s for you.This episode has been sponsored by Meter. Go to meter.com/aone to book a demo now! You can support the show at the link below.Support the showFind everything AONE right here: https://linktr.ee/artofneteng

Send us a textNetwork engineers don’t tune into corporate all-hands because they’re “lazy,” they tune out because the message often isn’t for them. In this episode of The Art of Network Engineering, Andy Lapteff sits down with longtime industry leaders Scott Robohn and Mike Bushong to unpack the disconnect between engineering teams and executive communications, and how to fix it.They talk about:Why engineers roll their eyes at town halls, earnings calls, and “four pillars of excellence”How leadership actually thinks about growth, stock price, cost centers, and enablementThe two jobs every company really has: build stuff or sell stuff, and where networking fitsHow to pitch your ideas in business terms so they get fundedWhy AI networking and data center infrastructure are the next durable growth areas for network prosThe difference between being part of the product vs. enabling the product,  and why it matters for your careerIf you’ve ever thought, “Just let me do my job,” this one’s for you. You’ll walk away knowing how to connect your automation, operations, or data center work to the outcomes your company actually cares about: revenue, speed, customer experience, and risk.Listen in, take notes, and then go advocate for your work like it matters, because it does.This episode has been sponsored by Meter. Go to meter.com/aone to book a demo now! You can support the show at the link below.Support the showFind everything AONE right here: https://linktr.ee/artofneteng