Diana's Tumblelog

Silicon Valley native. Puppy parent. Annual skateboarder.

[dianas]

The GTM Stack Anthropic Uses From Its Head of Industries: Surprisingly Familiar Names. Used In New Ways.

You’d assume Anthropic, of all companies, would be running on some self-grown, cutting-edge AI-native sales platform. Maybe a Salesforce killer they built themselves. Maybe Claude baked into …

[dianas]

Today’s leaders face increasing pressure on all sides, and their stress levels are higher now than they were even at the peak of the pandemic. Though stress can sharpen performance briefly, over time it erodes judgment, narrows perspective, and increases the risk of costly missteps. Most leaders have distinct default responses to it. This article outlines the six most common patterns: the calm lighthouse, the reinvention-oriented alchemist, the action-driven firefighter, the disciplined stoic, the relationship-focused diplomat, and the control-driven container. Each style has both strengths and blind spots that pressure can amplify. Leaders can increase their ability to perform under duress by identifying and understanding their default responses and then deliberately expanding their range of reactions—using simple tactics to regulate themselves, share the cognitive load, and alter their style in real time as conditions change.

[dianas]

Compound Engineering Gets an Upgrade

The AI-native engineering philosophy has expanded from four steps to eight

The short answer: Yes. Think of an AI workflow like a sandwich—the model is the workhorse filling, and we’re the bread, providing framing and taste.

Ideate → brainstorm → plan → work → review → polish → compound → repeat

Play to your strengths. Kieran’s compound engineering framework breaks the engineering workflow into four steps: Plan, work, review, and compound. AI takes care of the doing phase. “LLMs are very good at just following steps, doing deep work, working for hours or days, even now,” Kieran says. What’s left for flesh-and-blood humans are the steps before and after—the planning, where you frame the problem, and review, where you determine whether the output feels right (the bread!).

Humans can identify multiple solutions to the same problem—AI struggles at this. If your knee hurts, you could take Advil, stretch your IT band, or stop running on hard surfaces. Humans are good at diagnosing a problem from many different angles, an exercise agents struggle with, Dan says.

Taste is the final layer of bread. Once AI has done the work, the most important thing you can do is judge whether the output approaches the vision in your head. Does the output feel right—and if not, how can you reframe the problem until the AI produces something that does? This is what separates art, which has a point of view, from generic slop.

[dianas]

Skill Distillation

How a personal AI agent built on markdown skills lets a frontier model teach smaller, local models to do real work, without retraining.

This is fundamentally different from classical knowledge distillation, which compresses a big model’s soft probability outputs into a smaller model’s weights. It’s different from instruction tuning, which bakes behavior into weights through prompt-response pairs. It’s different from RAG, which retrieves facts.

Skill distillation retrieves procedures. The smaller model doesn’t have to know how to evaluate a company. It just has to know how to follow the steps.

[dianas]

From Vibe Coding to Spec-Driven Development | Towards Data Science

A 4.5-hour journey from idea to working fitness app with LLM agents

[dianas]

Judgment is the skill that matters most in the AI era

The most valuable AI skill isn’t prompting. It’s knowing when to push back.

[dianas]

Plastic User Interfaces

Salesforce and others are going headless. AI is making the user interface plastic — malleable to whatever the user needs, when they need it. The head isn't disappearing, it's becoming formless.

[dianas]

How Anthropic Rebuilt Its Sales Org From Scratch When Demand Went Vertical: 54% of New Enterprise Logos Now Come Self-Serve

When Claude Opus 4.6 shipped in December 2025, Anthropic’s commercial team came back from winter break to find demand had gone vertical. They hadn’t hired for it. They hadn’t plan…

[dianas]

In periods of rapid change, the teams that outperform everyone else are not those with the best plans or the most talent but those that learn the fastest. Research across thousands of teams reveals a consistent pattern: High-performing teams—“superteams”—build cultures of continuous improvement. Their leaders encourage experimentation even when things are going well, make curiosity and intellectual humility contagious, surface problems early, stay close to the work, give feedback that supports learning rather than punishing mistakes, and invest in people’s growth even when it doesn’t pay off immediately. When work is tied to shared meaning and progress matters more than perfection, teams become more resilient, adaptable, and capable of sustained success—in business settings and beyond.

[dianas]

How to use the Fish Model to discuss the Goal of a Transformation

How it helps to talk about the “Why” of the Transformation.

[dianas]

The power of vulnerability

Brené Brown studies human connection -- our ability to empathize, belong, love. In a poignant, funny talk, she shares a deep insight from her research, one that sent her on a personal quest to know herself as well as to understand humanity. A talk to share.

[dianas]

As AI accelerates product development and expands marketing’s responsibilities, most marketing organizations are struggling to keep up because their operating model—sequential, siloed, and coordination-heavy—hasn’t changed. The solution is a new structure built for human-agent collaboration, centered on a “brand code”: a machine-readable knowledge base encoding brand strategy, customer insights, and business rules that both people and AI agents can act on. Layered systems of specialized agents can then handle content creation, experimentation, distribution, and reporting at scale, while marketers shift from execution to strategic direction and judgment. Success requires rethinking not just technology but how organizations hire and develop talent—prioritizing people who can think in systems and shape how the platform evolves.

[dianas]

How I Use AI as a Product Manager

A practical map to the highest-leverage uses of AI in product management

[dianas]

How to respond when you’re put on the spot and not ready

What to say when your mind goes blank in a meeting

[dianas]

Our Uncertain Uncertainties

Even the experts inventing AI don’t know what will happen next. Is artificial general intelligence even possible? Can scaling continue? Will we need massive compute centers to make AI, or can we do it with a mere 25 watts like … Continue reading →

we should examine this direction carefully, and maybe give it a name: The Age of Ambiguity.

[dianas]

Emotional Intelligence Is A Life Skill

EQ and the Future of Work As AI creeps into the working world – either as an assistant to streamline current jobs or displacing others entirely – it’s becoming clear that leaders in the future will need to rely on a familiar skill set: emotional intelligence (I’ll call it EQ for short). While AI can

[dianas]

Leaders are making a choice with their AI strategy: Are they primarily seeking to improve the bottom line through automation and headcount reduction, or grow the top line in innovative ways through augmentation? As they make this decision, leaders are underestimating how employee perception—and the predictable behavioral dynamics that follow—will determine the success of their AI strategy. While automation strategies will likely show early gains relative to the deeper investment required for augmentation, but that augmentation will likely perform better in the long run. That’s because while automation offers immediate cost-savings, a company’s long-term success is determined by how people feel about their work, whether they meaningfully engage with new tools, and whether top talent stays.