Data Science Expert (AI Training Waitlist)

What you'll do

The ad's own example is a fair test of fit: a model can explain how an A/B test should be set up and still miss the leaked feature that makes the result look too good. If you have shipped a model and later found that kind of leak, you are who this network is looking for.

Sovrano's data science listing is a waitlist. After you apply you complete an AI interview about your field, about 20 minutes long, and then you wait. Sovrano checks the list from time to time and contacts people whose background matches an open project. There is no project behind the listing itself, the timing depends on demand, and some applicants will never be contacted.

On a project, the tasks would come from this set:

• writing realistic analysis and modelling tasks, with the worked answer an experienced practitioner would give
• reviewing other data scientists' work against an acceptance checklist
• grading AI answers on statistics, experiment design and code, including code that runs cleanly but answers the wrong question

Every piece is reviewed, and you are paid for what is approved.

Who fits

Sovrano accepts either hands-on data science or applied statistics experience, or a relevant degree that is in progress or finished. The skills it lists are Python, SQL, statistics, machine learning, experiment design, feature engineering, model evaluation and data visualisation.

The ad's examples lean toward judgment over syntax: leakage, flawed methodology, an analysis that answers a different question from the one asked. Practitioners who have reviewed colleagues' notebooks or sat in experiment reviews will find the work familiar. The usual basics also apply: fluent written English, a stable connection, a quiet place to work, and acceptance that some projects forbid AI assistance.

What it pays

Up to $150 an hour, which is the ceiling rather than a rate you are promised. Projects set their own rate and send it in writing with the hours and scope before work begins; the listing publishes no floor. Payment terms are not published on the listing.

Nothing is scheduled until a project matches your background.

Worth knowing

Good:

• Experience or a degree in progress both count, so working analysts without a statistics degree can apply
• The ad describes real reviewing work (leakage, methodology), which suits experienced practitioners
• Offers come with scope, hours and rate in writing, and you can decline
• Remote and part-time across North America, Europe and Turkey

Less good:

• Joining the waitlist does not mean you will be offered work
• Pay is a maximum with no stated minimum
• Only approved work is paid
• Hours per week are not given

About this listing

Posted by Sovrano on 28 October 2026 and confirmed open on its signed-in job list on 1 November 2026. The location is stated as regions, not a country list. See Sovrano AI.

How to get hired

• Screening is the roughly 20-minute AI interview the ad mentions; Sovrano's site says its assessment tests domain knowledge and evaluation skill.
• Put shipped work first on your profile: a model or experiment you took to production and how you validated it, with the stack (Python, SQL) named. Hands-on experience is the first requirement in the ad.
• Have one concrete example ready of a result that looked too good and what you found when you dug in, since leakage is the ad's own example of the work.
• Our interview prep section covers AI screening interviews.