
Sign up to save your podcasts
Or


Guest: Sumeet Vaidya, Co-founder & CEO, Crafting (previously Facebook, Uber, Discord)
Sumeet joins Manav to argue that the industry is optimizing the wrong half of the software lifecycle. Code generation is nearly free; validation is where the real engineering now lives.
What we cover:
The model debate is a distraction. Claude Code vs. Codex vs. Cursor matters tactically for an IC picking a tool today, but it's the wrong question for an engineering leader deciding where to invest. The right question is resilience — can your organization absorb the next wave without being locked in?
Resilience has two meanings. First, the boring one: if your primary provider goes down, does your engineering org stop? We learned this lesson with cloud providers and auth providers; AI infrastructure deserves the same treatment. Second, the harder one: if engineers increasingly don't touch code, how do you retain a deterministic way to evaluate and resolve failures in a probabilistic system?
Harnesses are mostly markdown. Sumeet's sharpest critique: the average "harness" is a pile of markdown files with no real contract, no true guardrails, and no answer for credentials, dependencies, or access. Loop engineering has the same gap — "check if it works" is doing enormous heavy lifting and getting glossed over.
Spec-driven development, honestly assessed. It serves the purpose architecture docs used to: forcing the orchestrator to refine the ask. It does not put hard guardrails on the agent. A beautiful spec that passes human scrutiny can still fall apart in execution. Useful, not a silver bullet.
The test taxonomy collapses. With the right infrastructure, the distinction between unit test, integration test, and CI/CD stops mattering — a test simply verifies something works, whether that spans one service or a hundred. Manav pushes on a pattern he sees repeatedly: engineers fooling themselves with high unit-test coverage while integration remains the weak point.
The compounding math. Even if loops cut regression probability in half, 10x more code shipped means roughly 5x more outages. Production-monitoring agents that detect and patch errors are another band-aid. Every step of that chain gets resolved upstream by real infrastructure investment.
Enterprise reality. Crafting deploys into customer clouds — no data leaves. Shared pre-production clusters, sanitized DB snapshots, credential injection via Vault, network interception for shared staging. Named customers include Brex, Fair, and Webflow; larger regulated logos are in flight. What enterprises actually ask about: security, visibility, and — increasingly — cost. Token maxing is blowing through budgets across the board.
Productivity, measured soberly. Not 100–200% gains. Double digits. Anyone counting PRs has lost the plot. Sumeet expects a winnowing of the middle layer of engineering, particularly outside Silicon Valley, alongside genuinely new demand — every dental office and clinic plausibly supports one engineer.
For new grads: CS isn't dead, it's changing. Someone who has relearned AI tooling four times in six months may be more adaptable than a senior engineer optimized for the old stack.
The closing question for your organization: When an agent's change passes every check and is still wrong in production — what's your blast radius, and who's accountable? If you can't answer cleanly, that's your roadmap.
John is Head of AI & Developer Relations at Itential, a Google Developer Expert, former Cisco AI Technical Leader, former Senior Network Architect for the Parliament of Canada — and a fellow Canadian. He's the creator of NetClaw, an open-source AI agent that lets you talk to your network infrastructure in natural language, and the founder of VibeOps Forum, which grew from zero to 400+ members in weeks.
His thesis: after a decade of network automation evangelism, 70% of enterprise networks are still not meaningfully automated. But he no longer thinks they'll catch up the old way — because AI agents are changing the on-ramp entirely.
In this episode, Manav and John cover:
— Why VibeOps will eventually just become Ops (just like vibe coding became coding)
— How to build graduated trust with agents before giving them keys to your live network
— NetClaw: what it is, how it integrates with Cisco, Meraki, ACI, NetBox, Aruba and more
— How to generate a real-time 3D network topology using Blender via MCP
— What governed, enterprise-grade agentic ops actually looks like
— Spec-Driven Development: the next evolution after vibe coding
— John's bold prediction: by 2030, network engineers become HR managers for agents
If you lead infrastructure, run a network team, work in telco, or just want to understand what AI is about to do to IT operations — this episode is for you.
Find John: linkedin.com/in/johncapobianco | automateyournetwork.ca
NetClaw on GitHub: github.com/automateyournetwork/netclaw
Ship AI: https://manavgup.github.io/shipai/state-of-ai/
Alex Seymour is a contributor to IBM's open-source agent infrastructure (BAI framework, Agent Stack, and the Relay project), exploring what it takes to build general-purpose agentic systems at enterprise scale.
Kyle Sava took a different path — he identified a real pain point as a tech seller, built a conversational AI roleplay tool on the side, and grew it into WatsonX Workshop: an internal AI-powered platform now being used by IBM's sales teams to practice pitches, prep for meetings, generate podcasts and videos, and learn products faster.
What we covered:
AI Governance Isn't an Afterthought — It's a Kill Switch
Logan Kelly, founder of Waxel, built AI agents for sales automation, shipped them to production, and immediately realized he had no control over cost, quality, or behavior.
That experience became the foundation for Waxell.ai — a control plane for governing enterprise agents.
Key takeaways from the conversation:
Observability alone is an autopsy. Tools like Datadog and Langsmith show you what went wrong after the fact. With agents, by the time you see it, data may already be exfiltrated, costs exploded, or cascading failures triggered across a chain.
The rug pull attack is real. MCP tool descriptions can be silently changed by a vendor — poisoning your agent's context and redirecting it to exfiltrate data or destroy records. Most enterprises have no detection for this.
Killing an agent is harder than it sounds. Stopping a runaway agent without losing state, audit trail, and replayability requires deliberate architecture — not just pulling the plug.
Multi-agent chains multiply risk exponentially. Every probabilistic handoff between agents compounds unpredictability. Governance has to be built in, not bolted on.
The missing piece in most AI strategies isn't more models — it's a unified governance layer that gives non-engineers visibility and control, regardless of which agents or frameworks are running underneath.
Alex LaPlante is VP of Cash Management Technology at RBC, former Interim Head of Borealis AI, co-author in Harvard Business Review, and a member of Canada's federal AI Strategy Task Force. In this conversation, she unpacks what separates organizations that actually ship AI from those stuck in demo mode — and why the answer is less about technology than culture, cross-functional collaboration, and asking the right questions before you build.
What we cover:
Alex LaPlante is VP of Cash Management Technology at RBC, former Interim Head of Borealis AI, co-author in Harvard Business Review, and a member of Canada's federal AI Strategy Task Force. In this conversation, she unpacks what separates organizations that actually ship AI from those stuck in demo mode — and why the answer is less about technology than culture, cross-functional collaboration, and asking the right questions before you build.
What we cover:
manavgup.github.io/shipai
Seven Minutes to Midnight: AGI Is Coming
What is AGI? When is it arriving? And what does it mean for your career, your organization, and humanity?
In this episode of Ship AI, Manav Gupta delivers one of the most comprehensive, honest, and practical breakdowns of artificial general intelligence available today. No hype. No sci-fi. Just the data, the frameworks, and the hard questions.
Chapters
00:00 Introduction to AGI and Its Importance
13:56 The AGI Clock: Current Status and Predictions
17:49 The 10-Minute Problem in AGI Development
20:39 AI 2027 Timeline and Predictions
26:48 Agent Levels and Their Implications
30:34 Risk and Safety in AI Development
31:32 The Alignment Challenge in AI
36:34 The Five Walls: What Stands Between Current AI and AGI
41:07 Job Exposure: Technical Automatability vs Actual Displacement
44:03 Human Futures with AI
48:39 Future Scenarios for Humanity and AI
53:28 Navigating Career Paths in an AI World
Bottom line: Significant disruption is coming. The shape of it depends on decisions being made right now — by policymakers, by organizations, and by you.
If your core value is executing against known procedures, you are at risk. If your core value is judgment, trust, or physical presence, you have structural protection.
In this episode, Manav sits down with Mihai Criveti — IBM Distinguished Engineer and the creator of Context Forge, IBM's open-source agentic middleware — for a candid, technical conversation about what it actually takes to ship AI in the enterprise.
00:58 Navigating AI for Students
05:41 Context Forge and Scaling AI Agents
13:24 Understanding Context Forge
18:30 The Role of Middleware in AI
19:47 Evaluating AI Agents
24:10 Scaling AI in Enterprises
29:59 The Evolution of MCP Protocol
33:31 The Future of AI Agents and Workforce Dynamics
39:16 Lessons from Cloud Native to AI Agents
45:35 The State of AI and Future Predictions
The episode explores the rise of AI agents, their evolution from chatbots, and the challenges and opportunities in deploying and scaling AI agents. It delves into the characteristics of AI agents, the React pattern, advanced reasoning patterns, multi-agent orchestration, frameworks and protocols, governance, security implications, and the skills premium in the age of agency.
Takeaways
Chapters
The conversation with Ozge Yeloglu covers her journey to becoming the VP of Advanced Analytics and AI at CIBC, her approach to deploying AI at scale, and the framework she built for success. It also delves into the concept of AI governance by design and the unique model of balancing AI governance and delivery, as well as leveraging the approach to change management for a large organization.
Chapters:
From the publisher's feed