BraintrustRating & model evaluation
Conduct & Communications Surveillance Analyst – AI Evaluation
up to $90/hr
About this role, in Braintrust's words
We’re looking for someone who has investigated trade or communications alerts and written up the findings. You’ll use that experience to create realistic cases that test whether AI can piece together what happened, identify the relevant evidence, and reach a defensible conclusion.
What you'll do
- Create realistic surveillance cases using a fictional firm’s policies, employee communications, and trade or order records.
- Write the finding you would document as an analyst and identify the evidence that supports it.
- Design cases that test whether AI can distinguish potential misconduct from legitimate activity, including when the records are incomplete or support more than one explanation.
Candidate background
What this role asks for
- Regulatory Compliance
- Risk Management
- Trade Surveillance
- Communications Surveillance
- Financial Compliance
Read from the posting's own words: show the exact sentences
- Regulatory Compliance: “listed by the platform under required skills: Regulatory Compliance”
- Risk Management: “listed by the platform under required skills: Risk Management”
- Trade Surveillance: “listed by the platform under required skills: Trade Surveillance”
- Communications Surveillance: “listed by the platform under required skills: Communications Surveillance”
- Financial Compliance: “listed by the platform under required skills: Financial Compliance”
Quoted from selected sections of Braintrust's posting. The full posting may state other conditions. Read the full posting
A closer look
More context, when you need it.
Full job descriptionThe original posting, in Braintrust's words
Posted by Braintrust, reproduced here so you can judge the role before clicking. Original posting ↗
We’re looking for someone who has investigated trade or communications alerts and written up the findings. You’ll use that experience to create realistic cases that test whether AI can piece together what happened, identify the relevant evidence, and reach a defensible conclusion.
What you’ll do
• Create realistic surveillance cases using a fictional firm’s policies, employee communications, and trade or order records.
• Write the finding you would document as an analyst and identify the evidence that supports it.
• Design cases that test whether AI can distinguish potential misconduct from legitimate activity, including when the records are incomplete or support more than one explanation.
What you’ll bring
• Hands-on experience triaging trade or communications alerts and documenting case findings. Relevant backgrounds include surveillance, business operational risk, and second-line investigations.
• Knowledge of market abuse rules, such as EU or UK MAR, and how potential insider dealing or market manipulation, including spoofing, can show up in communications and order data.
• Ideally, five or more years of relevant experience. Ten or more years with continued hands-on work is a plus.
• Experience with tools such as NICE Actimize, Nasdaq Trade Surveillance, Behavox, or Global Relay is a plus. So is experience with insider-risk or data-loss investigations, or writing and reviewing AI evaluation cases or training data.
• Professional English is required. Ability to work in either Korean or Arabic is preferred
Contract details
Remote freelance contract, approximately 20 hours per week, with an initial term of three months.
Published location regions: Europe
Working with BraintrustPay, getting started and work, from Braintrust's own pages
What the platform publishes in general. These are not extra terms confirmed for this particular role.
- Getting paid
- Through a processor such as Stripe or Wise. On hourly contracts you invoice through Braintrust and payment starts within 30 days, or later if the client has not paid. Its AI training page says gig tasks are paid weekly. usebraintrust.com/site-service-fees-terms, usebraintrust.com/ai-training
- Before your first paid task
- A profile and a short AI interview, usually under an hour, then ID verification and skills validation. Certification takes 1 to 2 days, and matching starts after it. usebraintrust.com/for-talent
- How much work
- No guaranteed minimum or regular pay. Its terms say Braintrust does not introduce talent to clients or find projects for them. usebraintrust.com/talent-terms, usebraintrust.com/terms
- Contract
- Independent contractor. Each project is a contract directly between you and the client, under terms with Freelance Labs, Inc. usebraintrust.com/talent-terms, usebraintrust.com/terms
Read on the pages linked above on 23 Sep 2026. Platforms change these terms, so check the linked page before you rely on one.
How does the pay compare?Across Rating & model evaluation roles, not this role alone
From our own daily reads of every open listing on the platforms we follow, not from Braintrust's posting. How we count.
Pay across the field: Rating & model evaluation
Across 156 roles in this field with a stated hourly rate in US dollars, on 19 platforms, the middle half pays $30 to $75 an hour at the top of the range, and the median is $45. This role's top rate, $90, is in the top quarter. Of the 69 distinct rates we count, 56 are lower and 11 are higher.
| Platform | Roles | Top of range, per hour |
|---|---|---|
| Meridial (Invisible) | 44 | $15 to $65 · median $30 |
| micro1 | 28 | $15 to $140 · median $45 |
| Prolific | 18 | $20 to $216 · median $40 |
| Alignerr | 11 | $120 |
| Innodata | 11 | $15 |
Sources and listing historyWhere this information comes from
This listing's history
- First seen here on 20 Nov 2026, and still on Braintrust's list when we last read it on 21 Nov 2026.
- The listed pay has not changed since then.
- Posted by Braintrust
- First seen by AIPayList
- Last seen live
- Listing ID
- 18199
Pay as the source states it: “USD 70–90 hourly”.
Platforms sometimes change a rate after posting. Every listing here is re-read at least once a day and corrected when its source moves, so a difference you see on the platform is usually a change made since the last read. How freshness works.
Keep exploring
More like this
Related field, different requirements. Check each role's own page before you apply.