Applied AI Fellow

Job Description

Applied AI Fellow

\n

\n

Kadre Labs · Fall/Winter 2026 Fellowship · Remote (US) · Part-time · 3 or 6 months

\n

\n

THE SHORT VERSION

\n

\n

You are an MBA student or recent MBA with real depth in sales, marketing, revenue operations, or go-to-market. You have watched AI reshape your function from the outside and you want to be the person building it.

\n

\n

This fellowship puts you inside live client engagements. You will find the process that is costing a business money, design the AI agent that fixes it, build the prototype yourself, and stay with it until it produces measurable results. You get paid a monthly stipend while you do it, and you share in the revenue from what you build, including after your fellowship ends.

\n

\n

WHY THIS ROLE EXISTS

\n

\n

Every company is about to employ more AI than people, and almost nobody is set up to run it. Systems integrators organized themselves around platforms: a Salesforce practice, an SAP practice. AI agents do not configure platforms. They automate processes and drive outcomes, and processes live in the client's business.

\n

\n

So the scarce skill is not writing code. It is knowing how a revenue org actually works, where the friction is, and what "better" is worth on the income statement. If you have run demand gen, carried a number, owned a funnel, or fixed a broken handoff between marketing and sales, you already have the expertise. What you need is the agentic AI stack, and that part we teach you.

\n

\n

WHAT YOU'LL DO

\n

\n

• Embed with a client team and map how a business process actually runs, including the parts nobody documented.

\n

• Break that process into discrete jobs an agent can hold, and write the spec for each one. Agents do exactly what you said, so the work is saying it precisely.

\n

• Identify where an agent creates measurable value, and defend the number to a CFO.

\n

• Build the working prototype yourself, going from idea to demo in days. You will use AI coding assistants, rapid app builders, workflow automation platforms, and process mapping tools. You are not expected to arrive knowing them. Most of our team learned on the job, and getting hands-on is part of what this fellowship is for.

\n

• Hand the validated prototype to our engineering team, who build the production version. You prove what the agent has to do and that it works. They make it production-grade, secure, and supportable. Nothing ships to production as a prototype.

\n

• Set the trust boundary: what runs unsupervised, and where a human signs off. You draw that line by asking what an error costs, how far it spreads, and whether it can be undone.

\n

• Judge the output. An agent can be fluent and wrong at the same time, and you will catch it, because you know what right looks like in this function.

\n

• Run discovery, demos, and onboarding, and become the person the client calls.

\n

• Track what the agent delivers against the client's P&L, tune it, and catch the failures that look fine on the surface.

\n

\n

WHAT YOU'LL GET

\n

\n

A monthly stipend. $2,000 per month for the length of your fellowship.

\n

\n

Variable compensation that outlasts the fellowship. Bonus eligibility plus performance-based revenue share. Agents you originate or oversee earn you a share of the revenue or profit they generate, and that share continues for a set duration after your fellowship ends. We put the terms in writing before you start.

\n

\n

A path to full-time. Fellows who deliver are our first call for full-time Forward Deployed Product Manager roles. The fellowship is how we hire.

\n

\n

The skill set the market is short on. Hiring for agent work has converged on a handful of capabilities: writing specifications precise enough for a machine, decomposing work into agent-sized jobs, evaluating output against a real standard, recognizing how agents fail, setting trust and escalation boundaries, and modeling what a deployment costs against what it returns. You will practice all of it on live client work, which is the only place it becomes real.

\n

\n

Real clients and real stakes. You will not be building demos for a drawer. Our clients are operating companies and private equity portfolio companies with real revenue on the line. We fund courses and certifications, and you will work directly with both founders.

\n

\n

FELLOWSHIP TERMS

\n

\n

• Length: 3 or 6 months. Tell us what fits your program and your schedule and we will work with it.

\n

• Commitment: part-time and flexible around coursework. We care about the client outcome, not your hours.

\n

• Location: remote, United States. Occasional client travel is possible.

\n

• Cohort: a small number of spots.

\n

• Work authorization: you must be authorized to work in the United States for the duration of the fellowship. We are not able to sponsor visas or support CPT/OPT arrangements at this stage.

\n

\n

WHO WE'RE LOOKING FOR

\n

\n

• An MBA student, or a recent MBA.

\n

• Real depth in sales, marketing, RevOps, or go-to-market. Manager level or above before business school is ideal. This is the background we most want to add right now.

\n

• Judgment about quality in your own function. You know the edge cases, the exceptions, and the answer that is technically right and practically useless. That judgment is the scarce input in agent work and it does not come from a model.

\n

• You write clearly enough to be taken literally. You can state a goal, its constraints, and its failure conditions without leaving room for interpretation.

\n

• You learn tools quickly. We do not care whether you already know the ones we use today. We care that you will teach yourself a new one to get a client unstuck.

\n

• You have been building with AI on your own time, whether or not anyone paid you to.

\n

\n

If you read that list and hesitated on one or two, apply anyway. We are hiring for judgment and drive. We would rather talk to you than have you screen yourself out.

\n

\n

HOW TO APPLY

\n

\n

Apply with your resume. No cover letter required.

\n

\n

A letter of recommendation is optional and welcome. If you have one from a professor, a manager, or a teammate who has seen you deliver something hard, include it.

\n

\n

We read every application and will get back to you either way.