Join the expert network
IQ256 works with independent human experts selected for specific decision blind spots. Engagements are short, focused, and briefed with a sanitized problem packet — you are asked a precise question, not handed a project.
This is independent contracting, not employment, and there is no guaranteed volume of work.
Who should apply
- You have operated, not only advised: decisions you made yourself, with consequences you lived through.
- Your judgement is concentrated — a defined domain where you are genuinely hard to replace.
- You are comfortable reasoning from an abstracted packet rather than full access.
- You will state a clear position, including one that disagrees with the client's premise.
- You can turn around a focused analysis in days, not months.
How working with IQ256 works
- Paid calibration
- A first paid calibration assignment establishes how you reason, how precisely you scope, and how you handle uncertainty. It is compensated work, not a free sample.
- Independent analysis
- You form your view before seeing other perspectives. You will not be asked to align with a predetermined conclusion.
- Conflict and confidentiality requirements
- Conflicts are declared before any packet is sent. Confidentiality terms apply to the packet, the question, and the existence of the engagement.
- No guaranteed volume
- Human experts are commissioned when a decision requires their specific contribution. We will not promise a pipeline that does not exist.
- Contribution, not only credentials
- Selection weighs what your input added that nothing else in the review could: a blind spot exposed, an assumption falsified, a trade-off made explicit. Unconventional backgrounds are welcome when they carry information others do not have.
- Capability assessment is ongoing
- Skills, availability, familiarity with a context, conflicts, and prices change. Fit is re-assessed per engagement rather than fixed at onboarding. There is no public ranking or leaderboard.
- Capability-card model
- The current capability-card model focuses on what an expert is useful for, which blind spots they expose, and what they are not the right source for. It does not require a public expert directory or a published network size.
We select for marginal information, not for seniority signals.
Contribution is evaluated on one question: what did the expert add beyond AI analysis and the other human experts already in the review? Volume of output is not the measure.
Tell us where your judgement is concentrated
This prototype does not submit anything. The button composes a structured email so you can review and send it yourself.
Nothing is submitted or stored by this website. Direct email to hello@iq256.com is the current route.