Design errors caught late in energy infrastructure projects trigger expensive redesigns and construction delays — but review teams are still checking documents manually, line by line. This role is to build an AI review assistant that flags inconsistencies, mismatched parameters, and checklist gaps across technical documents, making review teams faster and more systematic without replacing their judgment.
### What you'll build
A document ingestion and extraction pipeline that handles real engineering PDFs, not only clean digital text.
Cross-document comparison logic that surfaces mismatches in parameters, assumptions, references, or checklist coverage.
A reviewer-facing interface that ties every flag back to the exact page, section, table, or source snippet that triggered it.
A workflow that fits into how engineering review teams actually pressure-test projects.
### What you'll do
Own the product from document ingestion and structured extraction through discrepancy detection and reviewer experience.
Make pragmatic choices about OCR, layout parsing, citations, confidence, and escalation paths.
Focus the system on verifiable mismatches rather than overclaiming expert-level engineering conclusions.
Work with domain experts to understand what kinds of misses are expensive and what evidence they need to trust a flag.
Work with AI Fund's build team on wedge selection, technical scope, and the fastest path to real workflow value.
### What you need
Strong engineering skills and hands-on experience with document AI, PDF extraction, cross-document reasoning, or review workflows.
Comfort building systems for regulated or high-consequence environments where citations and traceability matter.
Judgment about how to combine deterministic extraction, heuristics, and model-based reasoning without overpromising.
The ability to build useful reviewer workflows instead of pure backend demos.
You know how to use AI coding assistants and modern AI tools to move faster while keeping the evidence chain explicit.
US work authorization. We are unable to sponsor visas for this role.
### Helpful but not required
Experience in AEC, utilities, legal tech, compliance tooling, or enterprise document review.
Experience with scanned PDFs, table extraction, document citation systems, or audit trails.
Founder or founding engineer experience in workflow-heavy B2B software.
### Who this is for
A builder who likes hard workflow products where accuracy, traceability, and user trust matter.
Someone who wants to solve an expensive operational problem with strong enterprise pain and clear ROI.
### What to know upfront
This is a 12-week, full-time, on-site residency in Mountain View, California.
Not every residency becomes a company. The goal is to pressure-test the idea quickly and honestly with real users and customers.
You will be building an AI Fund idea, not bringing your own startup idea into the program.
The process typically includes a Builder Event or equivalent working conversation, then a 48-hour Builder Challenge, then panel review with AI Fund build leadership.
The compensation is intentionally modest during the residency because the upside, if the idea works, is a founder-level role.
### Compensation
$10,000/month for 12 weeks ($30,000 total). This is a contract role during the residency. If the build leads to a funded company, the next step is a founder-level role with meaningful equity upside.
This is a strong opportunity for a builder who wants to take an idea from concept to working product alongside Andrew Ng and the AI Fund team. We are only considering candidates within commuting distance from Mountain View.
