Job Description

_Join as the 3rd person on the team. Founder-level impact + meaningful early
equity._

**We’re building ProNexus: the all-in-one platform for sourcing human
intelligence with AI.**

ProNexus uses AI agents to automate expert sourcing and vetting end-to-end. We
start with consulting and research teams who need niche experts for primary
research, then convert every project into structured data: identity
verification, fit signals, vetting outcomes, and post-call ratings. Over time,
this becomes an Expertise Graph—a proprietary, longitudinal record of expert
reliability that makes it faster to find the right people today and enables
frontier-quality human feedback for model training and evaluation tomorrow.

**About the role**
We’re hiring an Applied AI Engineer to design and build a Dataset Factory:
pipelines and tooling that transform expert interviews into evals, training
tasks, verifiers, and benchmark-grade datasets.

You’ll work with founders, domain experts, and (as we grow) research partners
to push model performance on tasks that matter in production—while maintaining
extremely high data quality standards.

**What you’ll do**

* You’ll work with founders, domain experts, and (as we grow) research partners to push model performance on tasks that matter in production—while maintaining extremely high data quality standards.
* Create grading rubrics and annotation guidelines that are consistent enough for teams to scale.
* Build verifiable tasks: programmatic checks, schema validation, constraint tests, and model-assisted graders where appropriate.

**Skills Required**

* Hands-on experience building LLM training data and/or eval datasets for frontier model teams, enterprise model deployments or data research / data annotation org producing datasets for model improvement

**Who you are**

* You’re a builder who loves shipping and learning fast.
* Think in first principles and turn ambiguity into execution
* Low ego
* Have high ownership and don’t need hand-holding
* Value speed + iteration and know when to go deep on quality
* Enjoy building with AI and can separate hype from what works (we consistently reevaluate our AI tooling on a monthly basis)
* Want to work hard (5–6 days/week) with a tight founding team

**Traction / momentum**

* 5 consulting/research firms currently in pilot
* Successful pilot + passed compliance + security review with a Big 4 consulting firm and in the process of onboarding
* Backed by advisors and angel investors with deep domain expertise (ex-BCG, ex-EY Partners)

*We have very high ambitions, and that requires an unusually strong team. We’re only hiring people we believe are truly top-tier. To set expectations clearly (and avoid wasting anyone’s time): This is a high-ownership role with a high bar. We give direct feedback early and often. If performance isn’t there, we won’t drag it out. Please don’t apply if you’re looking for a slow ramp or a low-intensity environment. Only apply if you're confident you can consistently ship at a high level and want to push yourself harder than you ever have before.