Steve Atherton

This is a really interesting article about writing agents with the Claude Agent SDK. It gets into design advice that you should consider when writing an agent with Claude. It chimes with what I’ve read about the development of Claude Code (write as little business logic as possible and concentrate on getting out of the way of the model). The general takeaway for me is from the three-block flow chart in the middle: Give the agent good tools/affordances in three categories: Gather Context, Take Action, Verify work. Do that at the right granularity, with good feedback, and then let the model do the driving. It’s deceptively simple, but it has a goldilocks quality to it. Trying to make a more complex flow chart with many more boxes and lines constrains the AI to just the things you could think of in advance and makes it overfit to what you know today. Trying to make it simpler by, for example, giving the agent a whole existing API as a single tool can overwhelm it with too much choice and then too much context in the responses. It’s definitely worth reading if you're interested in building or using agents, irrespective of AI vendor. https://lnkd.in/eKF2rXWb

7

How can you make sure orchestrating agents across many frameworks doesn't cost you a fortune? Today, Caelum Forder and I look into agent sessions from Coral Protocol.

16

CPTO Revolution: Why SaaS Companies Are Merging Product and Tech Leadership I’ve lived the pain of siloed leadership and the inefficiencies it creates many times. And I’ve also seen how tightly integrated product and tech can transform not just delivery, but outcomes. Traditional org charts are no longer keeping up. The most competitive SaaS companies aren’t hiring separate CTOs and CPOs anymore - they’re bringing both roles under one roof with CPTOs. This isn’t about headcount optimization. It’s about reducing friction and moving faster in a world where complexity is climbing and customer expectations evolve by the week. Here’s the core issue: when product and engineering leadership are separate, you’re always playing a game of telephone. Product defines what’s needed. Engineering estimates and negotiates. Product pushes. Engineering pushes back. Handoffs everywhere. Every step is a delay. A CPTO owns the entire journey - from insight to shipped product. One person accountable. One vision. One continuous flow. And it works. Companies running this model are seeing up to 40% faster go-to-market. Less tech debt. Clearer priorities. Higher team velocity. What Actually Changes 1. Aligned Metrics You stop optimizing for isolated outcomes. Product isn’t chasing adoption while engineering shields uptime. Both roll into shared business goals. 2. AI Gets Strategic With unified leadership, AI isn’t a bolt-on. It becomes core to how you design, build, and validate products - through data, feedback loops, and automation. 3. Developer Experience Matters Platform engineering gets the attention it deserves. Internal tools align with product delivery, and suddenly teams aren’t waiting - they’re shipping. UX maturity, platform architecture, technical scalability - all of it becomes a single continuum. You’re not just improving “process”; you’re aligning every decision to what users need and what’s feasible to build. The CPTO Skillset This isn’t a hybrid role - it’s an integrated one. You need the technical depth to guide architecture and the product judgment to say no to overengineering. You talk to engineers about latency and reliability, and to the board about market traction and differentiation. As AI accelerates software development, what matters is orchestration—of systems, teams, and strategy. You need to see the whole board. Cloud infra spend is growing ~10% yearly, mostly due to AI. That spend needs direction. CPTOs are the ones who can match innovation speed with platform efficiency. Companies clinging to the old CTO/CPO split are playing a slower game. Integration is no longer optional - it’s a competitive advantage. So—what’s keeping your org from making the leap?

15

The biggest hidden cost in telemetry are the blind spots. Missing signals, inconsistent collectors, and pipelines no one really owns can quietly distort what teams think they’re seeing. At scale, those small gaps turn into expensive mistakes. When you own the pipeline, you get clarity back on what’s collected, how it’s transformed, and where it goes. Understanding your data is a lot cheaper than assuming it. #OpenTelemetry #Observability #DevOps #Telemetry #DataInMotion #PlatformEngineering #OTel

13

In today’s Substack post from the Plan of Record series, I walk through the operational frameworks I’ve relied on to build high-performance product development teams and why they’re essential for diagnosing and fixing what’s broken. I also explain why we had to create the PoR Prioritization Framework to address a critical gap no other model could fill. Here’s the link: https://lnkd.in/ek9J_cqq #SaaS #ProductManagement #EngineeringLeadership #R&D #Prioritization #Execution #CPO #CTO #CPTO

14

I love this piece by my co-founder Scott Gerlach. Today StackHawk launched Business Logic Testing as part of our API Security offering. Today, most companies only run these critical test scenarios as part of their 1x year pen testing program, or if managed in house, humans spend a great deal of time producing these scenarios. Whether a pen test, or a manual test by your appsec team - they're happening infrequently and are expensive! With this new capability, AppSec teams can automate these tests on every application - every time they're deployed. Read more here: https://lnkd.in/gbBafgvq

36

3 Comments

A typical example of today’s shallow GenAI integrations, more marketing than meaningful help. The same goes for the “agent” buzzword we are overusing. If GenAI or an agent truly improves the experience, it should be obvious from the interaction. The best tech works quietly in the background, solving real problems without needing a label.

16

3 Comments