Manicule is building AgentRel: the systems that help companies become discoverable, cited, and recommended by AI agents.
For the last twenty years, companies optimized for people searching Google. The next interface is an agent answering a question, comparing products, or deciding what to recommend. We want to understand how those decisions happen and build the infrastructure companies need to influence them honestly.
That means solving two hard problems.
First: how does an LLM decide which content to retrieve, trust, cite, and use when recommending a company?
Second: how do we use AI to produce excellent technical content without filling the internet with slop
Ideally, you write excellent TypeScript, but you also understand the fundamentals of GEO—or have spent enough time studying LLMs to form a view on how they surface sources and recommend brands. You care about writing quality and can tell the difference between a useful technical explanation and fluent filler.
This is not a role where you receive finished specs and close tickets. You’ll work directly with our customers to understand the problem, decide what to build, and own it through production.
What you’ll do
Build the systems behind our documentation, technical content, and agent workflows.
Investigate how LLMs retrieve information, select sources, form recommendations, and cite brands. Lots of fun research to be done here.
Turn editorial judgment into agent workflows, evaluations, and quality checks.
Design how agents receive, retain, and use context across long-running jobs.
Work directly with founders, engineers, and DevRel teams at our customers.
Own projects from the first customer conversation through production.
Who we’re looking for
You have 3+ years of professional software-engineering experience.
You’re excellent with TypeScript and comfortable working across a product.
You have opinions on GEO and how LLMs discover and surface information.
You have strong opinions about what makes technical content useful, credible, and worth reading.
You build with agents yourself and think seriously about context, retrieval, evaluation, and human review.
You can operate without a detailed specification and know when to ask questions.
You want to talk to customers instead of treating customer context as someone else’s job.
You have obsessive attention to detail.
Your first 30 days
Probably the highest thing on my list is building an extremely durable and powerful context system.
Our agents need to gather context from dozens of sources, from a company’s Slack and Linear to GitHub and its docs, and use it to produce amazing work on the first try. These sources contain huge amounts of noisy, constantly changing information. You’d build, eval, and improve this production infrastructure.
Bonus points if you’ve worked in context management before.
Why now
We’re growing very fast - went from 0 to $30K MRR in 2 months (yes, actual real revenue). You’ll be early enough to have meaningful contribution to the actual product direction and lots of autonomy.
Also, we’re 2 chill young cofounders. You’ll like working with us.
Details
In person in San Francisco; remote may be possible during your first month.
