Subscribe
ChatGPT
ChatGPT
William Smith
William
CONVERSATIONS WITH CODE

The Context Curator: How Your "Soft" Skills Fuel the Next Generation of Business

If you’ve spent your career as a storyteller, journalist, or information architect, you’ve been training for the most important role in the AI era. In 2026, the most valuable part of any AI system isn't the code—it's the context.

When does a story stop being "content" and start being "infrastructure"?

We are currently in a moment where we get to design the roles of the future rather than just waiting for them to happen to us. For those of us who have spent years in podcasting, journalism, information architecture, or documentary filmmaking, there is a specific opportunity that is arguably the most interesting evolution of our craft. It’s called the Context Curator.

This isn't about learning to code; it's about realizing that your ability to listen, organize messy ideas, and find the "soul" of a story is actually the high-level technical architecture that AI systems are missing. You aren't starting from scratch; you are entering a "New Game Plus" version of your creative life where your pre-established skills are the primary assets.

What business problems does the Context Curator solve?

The Context Curator ensures that a company’s most valuable expertise is documented and utilized rather than lost. In most organizations, the specific knowledge that drives success is never recorded; it exists only in the minds of a few people or is scattered across unorganized digital messages.

Your role is to capture that expertise and make it available as a Contextual Object. By facilitating this capture, you ensure that the company’s history and voice are preserved and ready to be used as the primary intelligence for its AI systems.

What usually stops an AI from sounding like a brand?

The biggest hurdle for "stock" AI models like ChatGPT is that they are "Context-Poor". They can generate text, but they don’t know a brand's unique history, its hard-won failures, or the specific rants that define its culture. They sound like corporate strangers because they lack the "tribal knowledge" that lives in the heads of the team.

The problem isn't a lack of intelligence in the machine; it's a lack of memory. When you provide that memory, the story moves from being a piece of content to becoming the actual fuel that runs the business.

How do your existing skills translate into this new world?

You are already a master of the most valuable currency in 2026: Context. Your "soft" skills are actually high-leverage technical assets that allow you to move from making work that is "good enough" to work that is truly exceptional.

  • The Podcaster/Journalist: You use your ability to interview to perform "Knowledge Mining". You don’t just ask what a founder does; you extract the why—the nuanced definitions of "quality" or "vibe" that a machine would never find on its own.
  • The Information Architect: You take the messy "noise" of Slack threads and voice notes and structure them. You organize this disparate data so an AI system can use it in milliseconds, turning a folder of documents into a functional "Signal".
  • The Editor: A machine can transcribe an hour of tape, but it can’t find the "soul". You are the one who recognizes that when a leader says "speed," they actually mean "responsiveness". You bridge the gap between human intent and machine execution.

Building a "Personal Brain" with NotebookLM

As part of my own journey, I’ve been incorporating NotebookLM into my workflow, and I can tell you it has been incredibly useful for managing the "Cognitive Load" of complex projects. This is essentially where you build your "Personal Brain" or a "Living Brain" for a client.

NotebookLM works by allowing you to upload a massive variety of sources—YouTube transcripts, PDFs, Google Docs, and even your own AI chat histories—into a single, grounded environment. Unlike a standard chat, it doesn't just "know" things from the internet; it is locked specifically to the sources you provide.

The benefits of using this tool specifically for Context Curating are massive:

  • Slicing and Dicing Information: You can ask it to summarize 50 different interviews, find every instance where the founder mentioned "innovation," or generate a FAQ based solely on your internal notes.
  • Grounded AI Chats: Because it uses your documents as its primary source, the "hallucinations" are minimized. It cites its sources directly, so you can see exactly where an idea came from.
  • Operational Memory: It turns static files into a conversational partner. You can literally talk to your project’s history, making it the perfect tool for turning "tribal knowledge" into a durable asset.

What are "Contextual Objects" and why is RAG the real game-changer?

For years, we’ve used our skills to create beautiful artwork and assets that serve as the disposable face of a company. The Context Curator model takes those same storytelling skills and applies them to the technical infrastructure of the business through a process called RAG.

Retrieval-Augmented Generation (RAG) is essentially giving an AI a private library to consult before it speaks. Instead of the AI guessing based on its general training, it "retrieves" the specific Contextual Objects you’ve built—the transcripts, the manifestos, the project histories—and uses them to anchor its response.

By architecting these RAG-ready knowledge bases, you are creating an Operational Memory. You are building a "Living Brain" that allows an AI to:

  • Speak with an authentic brand voice because it’s literally reading your past work before it drafts.
  • Make decisions based on real history rather than generic corporate logic.
  • Eliminate "hallucinations" by citing actual company data for every claim it makes.

The stories you capture now aren't just for the sake of telling a story; they are the fuel that powers the RAG system, making the story the engine that finally runs the business.

Does this mean you need to become a "Prompt Engineer"?

Not exactly, but it does mean you are the one "stocking the pond" before the fishing even begins. Prompt engineering is an important skill—knowing how to talk to a model is essential—but even the most perfect prompt will fail if the AI is working in a vacuum.

Context Curating is the high-leverage work that happens before the prompt. While everyone else is fighting for attention in a crowded feed, you are architecting the reality the AI operates within. You are building the intelligence that makes AI actually useful to a specific business by providing the raw material it needs to be effective.

Who is built to succeed as a Context Curator?

This role isn't for everyone. It works best for the orchestrators and connectors—the people who naturally think in systems and possibilities rather than just isolated tasks.

If you are a generalist who enjoys figuring things out as you go, this is your leverage. You succeed here if you have a core craft—like video, audio, or information design—but you’ve always felt a pull to solve bigger, more structural problems. It’s for the creative who doesn't just want to "do the work," but wants to direct the work, fully using their curiosity.

What should you focus on first?

If you want to move into this role, start by treating your own work as a project.

  • Audit your own "Context": Gather your best writing, your saved voice notes, and your project history.
  • Build a "Personal Brain": Use a tool like NotebookLM to upload these samples and see how much better the AI performs when it actually knows who you are.
  • Practice "Extraction": Interview a peer about a specific success. Don't just record it—structure it. See if you can turn that conversation into a set of rules an AI can follow.

Why this matters more than speed

This isn't just about doing work faster; it’s about regaining control over what gets made. When you architect the context, you ensure the work points somewhere meaningful. The future isn't about replacing the storyteller—it's about the storyteller who knows how to give the machine a memory so that the story can finally run the business.

← Back to Digest
New Creative Roles February 13, 2026

The Context Curator: How Your "Soft" Skills Fuel the Next Generation of Business

If you’ve spent your career as a storyteller, journalist, or information architect, you’ve been training for the most important role in the AI era. In 2026, the most valuable part of any AI system isn't the code—it's the context.

William Smith
William Smith
The Context Curator: How Your "Soft" Skills Fuel the Next Generation of Business
Hyper-realistic podcast interview, vertical 1080x1920. Over-the-shoulder shot with blurred executive in foreground. Man in yellow sunglasses in sharp focus, wearing black cap, black over-ear headphones, grey hoodie, subtle stubble. Leaning toward a single desktop microphone stand (no boom arm), looking nearly into the camera. Modern corporate studio with floor-to-ceiling windows and soft, blurred city skyline. Shallow depth of field, cinematic natural light, soft rim light, high detail, professional color grading, premium commercial photography.

When does a story stop being "content" and start being "infrastructure"?

We are currently in a moment where we get to design the roles of the future rather than just waiting for them to happen to us. For those of us who have spent years in podcasting, journalism, information architecture, or documentary filmmaking, there is a specific opportunity that is arguably the most interesting evolution of our craft. It’s called the Context Curator.

This isn't about learning to code; it's about realizing that your ability to listen, organize messy ideas, and find the "soul" of a story is actually the high-level technical architecture that AI systems are missing. You aren't starting from scratch; you are entering a "New Game Plus" version of your creative life where your pre-established skills are the primary assets.

What business problems does the Context Curator solve?

The Context Curator ensures that a company’s most valuable expertise is documented and utilized rather than lost. In most organizations, the specific knowledge that drives success is never recorded; it exists only in the minds of a few people or is scattered across unorganized digital messages.

Your role is to capture that expertise and make it available as a Contextual Object. By facilitating this capture, you ensure that the company’s history and voice are preserved and ready to be used as the primary intelligence for its AI systems.

What usually stops an AI from sounding like a brand?

The biggest hurdle for "stock" AI models like ChatGPT is that they are "Context-Poor". They can generate text, but they don’t know a brand's unique history, its hard-won failures, or the specific rants that define its culture. They sound like corporate strangers because they lack the "tribal knowledge" that lives in the heads of the team.

The problem isn't a lack of intelligence in the machine; it's a lack of memory. When you provide that memory, the story moves from being a piece of content to becoming the actual fuel that runs the business.

How do your existing skills translate into this new world?

You are already a master of the most valuable currency in 2026: Context. Your "soft" skills are actually high-leverage technical assets that allow you to move from making work that is "good enough" to work that is truly exceptional.

  • The Podcaster/Journalist: You use your ability to interview to perform "Knowledge Mining". You don’t just ask what a founder does; you extract the why—the nuanced definitions of "quality" or "vibe" that a machine would never find on its own.
  • The Information Architect: You take the messy "noise" of Slack threads and voice notes and structure them. You organize this disparate data so an AI system can use it in milliseconds, turning a folder of documents into a functional "Signal".
  • The Editor: A machine can transcribe an hour of tape, but it can’t find the "soul". You are the one who recognizes that when a leader says "speed," they actually mean "responsiveness". You bridge the gap between human intent and machine execution.

Building a "Personal Brain" with NotebookLM

As part of my own journey, I’ve been incorporating NotebookLM into my workflow, and I can tell you it has been incredibly useful for managing the "Cognitive Load" of complex projects. This is essentially where you build your "Personal Brain" or a "Living Brain" for a client.

NotebookLM works by allowing you to upload a massive variety of sources—YouTube transcripts, PDFs, Google Docs, and even your own AI chat histories—into a single, grounded environment. Unlike a standard chat, it doesn't just "know" things from the internet; it is locked specifically to the sources you provide.

The benefits of using this tool specifically for Context Curating are massive:

  • Slicing and Dicing Information: You can ask it to summarize 50 different interviews, find every instance where the founder mentioned "innovation," or generate a FAQ based solely on your internal notes.
  • Grounded AI Chats: Because it uses your documents as its primary source, the "hallucinations" are minimized. It cites its sources directly, so you can see exactly where an idea came from.
  • Operational Memory: It turns static files into a conversational partner. You can literally talk to your project’s history, making it the perfect tool for turning "tribal knowledge" into a durable asset.

What are "Contextual Objects" and why is RAG the real game-changer?

For years, we’ve used our skills to create beautiful artwork and assets that serve as the disposable face of a company. The Context Curator model takes those same storytelling skills and applies them to the technical infrastructure of the business through a process called RAG.

Retrieval-Augmented Generation (RAG) is essentially giving an AI a private library to consult before it speaks. Instead of the AI guessing based on its general training, it "retrieves" the specific Contextual Objects you’ve built—the transcripts, the manifestos, the project histories—and uses them to anchor its response.

By architecting these RAG-ready knowledge bases, you are creating an Operational Memory. You are building a "Living Brain" that allows an AI to:

  • Speak with an authentic brand voice because it’s literally reading your past work before it drafts.
  • Make decisions based on real history rather than generic corporate logic.
  • Eliminate "hallucinations" by citing actual company data for every claim it makes.

The stories you capture now aren't just for the sake of telling a story; they are the fuel that powers the RAG system, making the story the engine that finally runs the business.

Does this mean you need to become a "Prompt Engineer"?

Not exactly, but it does mean you are the one "stocking the pond" before the fishing even begins. Prompt engineering is an important skill—knowing how to talk to a model is essential—but even the most perfect prompt will fail if the AI is working in a vacuum.

Context Curating is the high-leverage work that happens before the prompt. While everyone else is fighting for attention in a crowded feed, you are architecting the reality the AI operates within. You are building the intelligence that makes AI actually useful to a specific business by providing the raw material it needs to be effective.

Who is built to succeed as a Context Curator?

This role isn't for everyone. It works best for the orchestrators and connectors—the people who naturally think in systems and possibilities rather than just isolated tasks.

If you are a generalist who enjoys figuring things out as you go, this is your leverage. You succeed here if you have a core craft—like video, audio, or information design—but you’ve always felt a pull to solve bigger, more structural problems. It’s for the creative who doesn't just want to "do the work," but wants to direct the work, fully using their curiosity.

What should you focus on first?

If you want to move into this role, start by treating your own work as a project.

  • Audit your own "Context": Gather your best writing, your saved voice notes, and your project history.
  • Build a "Personal Brain": Use a tool like NotebookLM to upload these samples and see how much better the AI performs when it actually knows who you are.
  • Practice "Extraction": Interview a peer about a specific success. Don't just record it—structure it. See if you can turn that conversation into a set of rules an AI can follow.

Why this matters more than speed

This isn't just about doing work faster; it’s about regaining control over what gets made. When you architect the context, you ensure the work points somewhere meaningful. The future isn't about replacing the storyteller—it's about the storyteller who knows how to give the machine a memory so that the story can finally run the business.