2/18/26: AI For Operators

It's Time to Face Reality, 4 new jobs, 10 links

February 18, 2026

Hi there,

Welcome back to AI for Operators. Here’s what we’ve got for you this week:

  • The Essay: Operators: It's Time to Face Reality
  • The Jobs: 4 AI strategy roles
  • The Links: 11 curated reads
  • The Events: 2 upcoming

Operators: It's Time to Face Reality

By Tom Guthrie

If you haven't yet realized that the way companies work has changed, it's time to face reality. Off-the-shelf AI systems can now do a substantial portion of tasks that white collar workers do today, and the improvement curve is steep. Operators (chiefs of staff, business operations leads, program managers, RevOps professionals) are not immune to that change.

Today, as in right now, AI can generate a good first draft of your next board deck and write your next investor update. It can provide solid advice on tough personnel issues with a little context. It can create an internal web app to track and update OKRs. It can generate a full plan for an executive offsite, gather venue options, and email those venues for quotes. It can create and update a financial model in Excel. It can take product usage data, qualitative input from your sales team, and customer feedback and create a PRD for a new feature — then build that feature and push it to production.

You're likely already thinking: if these tasks are taken off my plate, what will my role be? Your answer might be: "I'll work with leadership to move our strategic priorities along faster." But large parts of their jobs will be done by AI, too. It will take longer for operators in government and highly-regulated industries to feel the impacts. At larger companies, these changes will take longer to take hold, too. Still, technological changes of this magnitude don't leave any organization untouched.

Indeed, there's a case to be made that many operational roles could disappear entirely, with leaders instead being supported by always-energetic, instantly responsive agentic systems that extend their capabilities without the back-and-forth any human relationship requires.

Fundamentally, operators exist for two reasons: because leaders can't be everywhere and make every decision, and because organizations of any real size experience coordination problems best solved by someone with broad context and significant authority. You can envision a company that's just a CEO perched atop a massive army of AI agents with complete transparency and absolute control. For small businesses, this may be possible. But for any organization that generates large amounts of data, requires many decisions, interfaces with the physical world, or requires sustained coordination with people outside the organization, human workers will still be part of the equation.

The most likely near-term future involves leaner teams with both human and AI workers. With AI handling more of the task-level work, operators can focus on two things: setting up and maintaining the systems that help their organizations perform at their peak, and doubling down on the work that only humans can do.

What should you do now? First, become the AI champion within your organization. That means continually learning about and experimenting with cutting-edge systems. It means leading by example and showing others what can be accomplished with AI. It means gaining buy-in to spread those systems throughout the organization, even if it means full-scale transformation. Being AI-first is no longer an optional part of an operator's job, it's the core of it.

Second, double down on the tasks only humans can do. That means cultivating curiosity, wisdom, and equanimity in the face of unsettling change. It means spending more time with people, listening to them, and learning from them. It means finding sources of information AI doesn't have access to - off-the-record conversations, channel checks, talent that hasn't been identified yet. It means becoming a better manager of both people and agents. It means becoming someone people want to be around.

Will all operators evolve to meet this moment? No. But the role will persist: increasing organizational complexity will require highly capable, cross-functional operators, and leaders still need trusted advisors with broad context. Steering organizations to become AI-first will require exactly the kind of people who read newsletters like this one.

Your to-do list for this week:

Take one recurring task from your workload and automate it

Block two hours to experiment with a frontier AI tool on a real work problem

Have a conversation with your leader about how AI changes your shared priorities and the priorities of the organization

Some AI strategy and implementation roles that caught our eye:

  • New York, NY
  • Comp: $143,000 - $210,000

AI Enablement Lead — Ridgeline

  • New York, NY
  • Comp: $170,000 - $190,000
  • New York City Metropolitan Area
  • Comp: $200,000 - $350,000

AI Innovation Manager — Fenwick & West

  • Washington, DC
  • Comp: $126,000 - $189,000

Practical

  • NotebookLM rolls out slide generation — Google's NotebookLM now automatically generates slide decks from your sources, turning research and notes into presentation-ready materials without the usual formatting grunt work. A potential time-saver for operators who spend hours translating documents and insights into stakeholder presentations.
  • How a CEO Uses AI as His Chief of Staff — A CEO shares how he's turned Claude into his personal chief of staff, using it to draft communications, prep for meetings, and think through strategic decisions—complete with specific prompts and workflows you can steal. If you're looking for practical ways to elevate your own use of AI beyond basic tasks, this thread is a goldmine of real-world applications.
  • AI Infrastructure for Orgs: Zulip as Shared Nexus for Humans and Bots | John Dean posted on the topic | LinkedIn — WindBorne Systems embeds AI agents directly into their team chat (Zulip), creating a shared context where engineers and non-technical teams collaborate with bots that evaluate case studies, monitor training runs, and audit code—an approach that requires serious infrastructure investment but enables a small team to move fast. The key insight: you can't buy your way to AI transformation with SaaS products; you need open-source tools and engineering capacity to build custom integrations that fit your actual workflows.
  • No Coding Before 10am | Michael Bloch — A startup's radical approach to engineering in the AI era: spend mornings defining problems and letting AI agents write the code, fundamentally shifting the role of engineers from coders to orchestrators. The playbook includes practical tactics for prompt engineering, agent workflows, and restructuring your dev team around AI collaboration.
  • The future of software engineering — ThoughtWorks gathered senior engineering leaders to explore how AI is reshaping software development, finding that while AI coding assistants boost productivity by 20-30%, the real transformation lies in how teams will need to rethink architecture, testing, and code review practices. The report offers practical frameworks for technical leaders navigating the shift from AI as a tool to AI as a collaborator in the development process.

Perspectives

  • Intelligent AI Delegation — A new framework for AI agent delegation goes beyond simple task-splitting to handle the messy reality of transfer of authority, accountability, and trust—critical for operators building systems where AI agents need to reliably work with each other and with humans when things don't go as planned.
  • The financialisation of AI is just beginning — Wall Street is racing to create financial products tied to AI compute capacity, from derivatives on GPU time to indexes tracking model performance—a shift that could fundamentally reshape how businesses access and budget for AI infrastructure. If you're planning multi-year AI investments, understanding these emerging financial instruments could unlock more flexible procurement strategies and better risk management.
  • Dario Amodei — "We are near the end of the exponential" — Anthropic's CEO warns that AI's exponential progress could plateau within 1-3 years due to data limitations, making this narrow window critical for operators to build competitive advantages before the technology commoditizes.
  • Anthropic’s Chief on A.I.: ‘We Don’t Know if the Models Are Conscious’ — Anthropic CEO Dario Amodei admits the AI industry's dirty secret: we're racing ahead with systems whose inner workings remain fundamentally mysterious, including whether they might be conscious. For operators deploying AI systems at scale, this candid acknowledgment highlights the critical importance of building robust testing and monitoring frameworks rather than assuming we fully understand what we're working with.
  • You've been kicked out of the arena, you just don't know it yet — It might be too late for you to transform your company into a winner with AI. To have a chance will require drastic, immediate change.
  • AI Won’t Automatically Make Legal Services Cheaper — Despite AI's potential to streamline legal work, three critical bottlenecks—regulatory barriers, liability concerns, and the complexity of actually deploying AI systems—mean we shouldn't expect dramatically cheaper legal services anytime soon. A useful reality check for operators tempted to assume AI will automatically slash costs in any professional services domain.

Thanks for reading,
Tom Guthrie