Conjuring digital companions: How I'm thinking better through AI

This last weekend I created another LLM-powered tool, Impersonaid (all puns intended). It’s a docs user simulator: you provide the URL of a document (or its Markdown source), select the virtual persona, and start a conversation about the content. Right after I released it, I realized that I had been talking to an imaginary friend to create more fictional interlocutors to interact with. It’s not as bad as it sounds, though. In fact, I would argue this is what writers are meant to do.

As a writer, I feel less alone and less powerless when working with an AI

Here’s an undeniable fact: technical writers, who are well equipped to detect product pains, are almost always left to fend for themselves, with little or no organizational support. Ours is a life of professional loneliness, occasionally broken by interactions with other writers with whom we share the burden, trying to compose together a magnum opus with no end in sight. Our audience is almost always unreachable, the impact of our work hard to measure. We’re hermits.

It’s in the cold warmth of AI responses that some of us are finding solutions. Capable as we are of crafting powerful incantations, we summon digital golems who help us build, fix, and test docs toolchains, draft content, or even validate it. Like desperate necromancers, we conjure artificial personas and coding sidekicks from the land of dead words. As bizarre as it might sound, I feel less alone when working with the help of a large language model.

Because, isn’t this what writers do, to reach for dictionaries, books, style guides, thesauruses, and other resources in search of answers and inspirations? And what is a large language model if not an animated corpus? We question it and it answers back, turning the simmering monologue of the writer into a conversation, our motions a written choreography, a fictional dialogue the likes of which writers have always arrayed in their notebooks.

LLM-powered tools are the greatest enablers for writers in tech

Our biggest challenge as technical writers is bridging the gap between the tech we document and the users trying to understand it. LLMs provide us with the support we need to build that bridge. But it’s more than just a bridge to the user; it’s a mirror for the self. Running a draft through an LLM allows me to remix, rewrite, and truly understand what I’m trying to say. I’m not mindlessly generating with AI; I’m thinking through AI. It’s the writer’s equivalent to rubber duck debugging.

The tools we build using LLMs also allow our words to materialize as direct product impact and socialize our thoughts inside organizations. Think of the typical debates over a confusing user interface. Before, I used to argue based on my own expertise and intuition. Today, I could say, “I ran our docs through Impersonaid and it got stuck at this exact step. Here’s the full transcript.” Suddenly, my argument becomes a reproducible linguistic experiment.

The other day I contributed code to a project at work, my Impostor Syndrome inside the trunk of my car, handcuffed and gagged. I could do this thanks to an AI-powered LLM. This, too, is an act of bridge-building: thanks to all I can automate and script without asking for dedicated devs, my documentation can get to users faster and in better shape. For technical writers, LLM tooling helps commoditize self-publishing. We can interface with the world on our own.

Now go and create shards of your own thoughts through LLMs

For the other hermits out there, staring at the screen and writing into the void, I can only say this: build your own strange friends. Assemble your own companions from code and prompts. We may have chosen a profession of solitude, but we are entering an era where we no longer have to work in isolation. In this bizarre new world, we have the power to create the very audience and collaborators we’ve always lacked.

And perhaps, in sharing these creations, we’ll find we are not so alone after all.