Skip to content

Research report ยท October 2026 edition

The Market for Math & STEM Work

On 31 Oct 2026, 248 roles in Math & STEM were open on 28 platforms (Invisible listed 75 of them (30.2%)); their median advertised rate was $40/hr, the same as the $40/hr of all AI training work; each open role drew 0.87x the apply clicks of the average role.

Read the month's story: The AI Training Job Market in October 2026

Provisional figures. This month is complete, but late data can still arrive until 3 Nov 2026, when the edition is frozen for good.

Data window
1 Oct 2026 to 31 Oct 2026
31 days, UTC
Computed
1 Nov 2026, 01:17 UTC
Final once frozen
Source
The public job boards of the AI training platforms this site tracks, read every hour. Apply clicks are made on AITraining.jobs; full descriptions come from the platforms whose boards publish them.
Method
Version 5. Definitions
CC BY 4.0, how to cite

Every figure on this page as CSV and JSON, sent to your work email.

Key findings

  1. 1

    28 platforms listed the 248 open roles in Math & STEM; Invisible held 75 of them (30.2%) and the three largest 142 (57.3%) (HHI 1,426, unconcentrated, against 522 for all AI training work).

    Figure 3
  2. 2

    227 of the 248 open roles in Math & STEM (91.5%) stated an hourly rate, with a median of $40/hr (typical pay $35 to $150), the same as the $40/hr of all AI training work.

    Figure 4
  3. 3

    Each open role in Math & STEM drew 2.97 apply clicks a week on AITraining.jobs, 0.87x the average AI training role: the 6th lowest of 12 kinds of work whose demand reads on its own.

    Figure 6
  4. 4

    43 of the 248 open roles in Math & STEM (17.3%) accepted applicants from any country, against 18.5% of all AI training roles; of the 20 countries compared, residents of the United States (145), then the United Kingdom and India (112 each) could apply to the most, and residents of Palestine to the fewest (55).

    Figure 9
  5. 5

    172 of the 248 open roles in Math & STEM (69.4%) name an expert field in their title, 14 expert fields in all; Mathematics and statistics (93), Physics and astronomy (40) and Data science and machine learning (11) were named most.

    Figure 10
  6. 6

    Of the 232 open roles in Math & STEM with a full description, 177 (76.3%) mention a PhD or master's degree, 10 (4.3%) a professional credential and 65 (28.0%) an assessment or test; the 20 stating years of experience ask for a median of 2 years.

    Figure 11
  7. 7

    Over the 31 days from 1 Oct 2026 to 31 Oct 2026, platforms listed 67 new roles in Math & STEM and removed 77, and a removed role had been listed for a median of 39 days (30 across all AI training work).

    Figure 2
  8. 8

    Within Math & STEM, roles listed during the window drew 8.34 apply clicks per role a week, against 0.67 for roles listed more than 12 weeks before the window.

    Figure 7

Figure 1

Open roles, week by week

Is the number of open roles in Math & STEM rising or falling?

WeekOpen roles (fixed panel)Index
1 to 4 OctPartial week
169
103.7
5 to 11 OctPartial week
163
100.0
12 to 18 OctPartial week
161
99.2
19 to 25 OctPartial week
162
102.5
26 to 31 OctPartial week
162
102.5

Sample. Average open roles in Math & STEM over each week's days, on the 18 platforms tracked since 1 Oct 2026 (the fixed panel; 12 of them list this work), 1 Oct 2026 to 31 Oct 2026.

The index is the panel's open roles against the week of 5 Oct 2026 (100), chained week to week, beside the same index for all AI training work on the same platforms.

This site changed how it reads 4 platforms' boards in this span (Alignerr on 14 Oct 2026; OpenTrain AI on 14 Oct 2026; Handshake AI on 14 Oct 2026; Prolific on 18 Oct 2026). From each change to the end of its week the roles that came into view are counted as already listed, not as new listings, and the roles that left view are not counted as removals; changes in open roles are measured like for like across it, on the roles that did not move with it. Each platform's own report says what changed.

SuperAnnotate (SME) (first tracked 10 Oct 2026), Turing (first tracked 10 Oct 2026), RemoExperts (first tracked 10 Oct 2026), Sovrano AI (first tracked 11 Oct 2026), Anyone AI (first tracked 11 Oct 2026), Vetto (first tracked 14 Oct 2026), Claru (first tracked 14 Oct 2026), Weekday (first tracked 18 Oct 2026), Rise Data Labs (first tracked 18 Oct 2026), Scale AI Human Frontier Collective (first tracked 18 Oct 2026), Anthropic (first tracked 21 Oct 2026), DataAnnotation (first tracked 22 Oct 2026), TELUS Digital AI (first tracked 22 Oct 2026), Aligned Labs (first tracked 22 Oct 2026), Surge AI (first tracked 22 Oct 2026) and Mistral AI (first tracked 25 Oct 2026) list this work but were added to the dataset later, so they are left out of the panel and counted only in the last column: a platform's arrival never reads as growth.

Every week is marked: the kind's change differs no more than chance from the rest of AI training work's, so its weeks do not read on their own.

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

Figure 2

Listings added and removed

How fast do roles in Math & STEM turn over, and is the stock growing on the same platforms?

WorkOpen on 1 Oct 2026Open on 31 Oct 2026, same platformsChangeNew listings
Math & STEMSmall sample175170
-2.6%
67
All AI training work2,6652,554
-4.2%
1,638

Sample. Roles listed from 1 Oct 2026 to 31 Oct 2026: 365 in Math & STEM, 7,396 across all AI training work.

Change compares the roles open on the first and the last day of the window on the platforms tracked when it opened, so a platform added during the window does not read as growth.

The kind's change is marked: it differs no more than chance from the rest of AI training work's.

A new listing first appeared in the window on a platform already tracked before that week; a platform's first week (its back catalogue) is not counted.

Median days listed runs from a removed role's listing date to its last sighting, for removed roles with a known listing date (at least 3; 72 here).

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

Figure 3

Who lists it

Which platforms list roles in Math & STEM, and what does each advertise?

PlatformOpen rolesShare of open rolesMedian per hour
Invisible75
30.2%
$35
Anyone AI39
15.7%
$40
AfterQuery28
11.3%
$115
Alignerr16
6.5%
$100
Rise Data Labs13
5.2%
$30
Handshake AI10
4.0%
n/a
micro19
3.6%
$150
OpenTrain AI7
2.8%
$40
Mercor6
2.4%
$50
SuperAnnotate (SME)5
2.0%
n/a
Turing5
2.0%
n/a
Weekday5
2.0%
$30
Sovrano AI4
1.6%
n/a
TELUS Digital AI4
1.6%
n/a
DataAnnotation3
1.2%
$80
Aligned Labs2
0.8%
n/a
Braintrust2
0.8%
n/a
Claru2
0.8%
n/a
Prolific2
0.8%
n/a
RemoExperts2
0.8%
n/a
xAI (SpaceXAI)2
0.8%
n/a
Anthropic1
0.4%
n/a
Mistral AI1
0.4%
n/a
Scale AI Human Frontier Collective1
0.4%
n/a
Surge AI1
0.4%
n/a
Terac1
0.4%
n/a
TransPerfect DataForce1
0.4%
n/a
Vetto1
0.4%
n/a

Sample. 248 roles in Math & STEM open at the end of 31 Oct 2026, by the platform that lists them: HHI 1,426 (unconcentrated), against 522 for all AI training work.

HHI: the sum of every platform's squared percentage share of the open roles, 0 to 10,000; under 1,500 is unconcentrated, 1,500 to 2,500 moderately and above 2,500 highly concentrated (US merger-guideline thresholds).

Shares are of listings, not of people hired or money paid: a platform posting one role per language or country weighs more here than its hiring may.

A median needs at least 3 roles stating an hourly rate.

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

Figure 4

What it pays

What do roles in Math & STEM advertise per hour, against AI training work as a whole?

WorkMedian per hourTypical pay fromTypical pay toStating an hourly rate
Math & STEM
$40
$35$15091.5%
All AI training work
$40
$30$12581.4%

Sample. Roles open on 31 Oct 2026: 248 in Math & STEM, 227 of them stating an hourly rate; 4,381 across all AI training work.

Pay is the hourly figure a posting advertises (the middle of a stated range), counted once per distinct rate as the job board counts it. The median and typical pay are the board's: typical pay runs from the 25th percentile of what roles pay (the middle of each advertised range) to the 90th-percentile advertised ceiling, stated from 3 distinct rates.

Roles paid per task or per project, or that state no figure, do not state an hourly rate.

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

Figure 5

How its rates are distributed

What share of roles in Math & STEM pay $50 an hour or more?

Advertised rateOpen rolesShare of priced rolesAll AI training work
Under $15/hr0
0.0%
6.9%
$15 to $25/hr2
0.9%
10.6%
$25 to $35/hr14
6.2%
11.9%
$35 to $50/hr121
53.3%
26.0%
$50 to $75/hr19
8.4%
14.7%
$75 to $100/hr26
11.5%
15.3%
$100 to $150/hr26
11.5%
9.4%
$150/hr or more19
8.4%
5.1%

Sample. 227 roles in Math & STEM open on 31 Oct 2026 that state an hourly rate, against 3,565 across all AI training work.

Each role counts once, at the middle of its advertised range. The bands are those of the pay-transparency report.

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

Figure 6

Applicant demand per open role

Do roles in Math & STEM draw more or fewer applicants per opening than the average AI training role?

RolesClicks per role a weekIndex
Math & STEM
2.97
0.87x
All AI training work
3.41
1.00x
Its roles stating an hourly rate
3.14
0.92x
Its roles stating no hourly rate
1.20
0.35x
Its roles open worldwide
1.60
0.47x
Its roles limited by country
3.26
0.96x

Sample. Apply clicks on a specific role on AITraining.jobs, from 1 Oct 2026 to 31 Oct 2026: 2,798 on roles in Math & STEM, against the days those roles were open.

Clicks per role a week = apply clicks / (open role-days / 7). Index 1.00 is the rate of every role on the site.

Math & STEM against every kind of work: Rows with fewer than 3.6 open role-days (0.5 role-weeks) are marked as a small sample and not ranked: below that the pooled rate of the kinds of work weighs more than a row's own clicks (the weight fitted on their own clicks).

Its roles by stated pay, against each other: Rows with fewer than 8.2 open role-days (1.2 role-weeks) are marked as a small sample and not ranked: below that the pooled rate of the two sides weighs more than a row's own clicks (the weight fitted on their own clicks). Its roles by reach, against each other: Rows with fewer than 15.3 open role-days (2.2 role-weeks) are marked as a small sample and not ranked: below that the pooled rate of the two sides weighs more than a row's own clicks (the weight fitted on their own clicks). All AI training work is one set, read from its first open role-day.

Apply clicks measure interest among visitors to AITraining.jobs, who see roles open in their country and recent, priced roles first. Clicks are never split by platform.

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

Figure 7

Applicant demand by listing age

Do newer roles in Math & STEM draw more applicants than older ones?

Listing ageClicks per role a weekIndex (this work = 1)
Listed during the window
8.34
2.81x
Listed up to 4 weeks before it
1.15
0.39x
Listed 4 to 12 weeks before it
13.82
4.65x
Listed more than 12 weeks before it
0.67
0.22x
Listing date unknown
0.18
0.06x

Sample. Apply clicks on roles in Math & STEM, from 1 Oct 2026 to 31 Oct 2026, by when each role was listed relative to the window's first day (1 Oct 2026).

Listing date: the date the job board gives for the posting, or else the first day the role appeared in the feed. A role that arrived with a platform's back catalogue and gives no date has none.

The index divides each row's rate by the rate of all roles in this kind of work, so it reads within the kind.

Rows with fewer than 1.3 open role-days (0.2 role-weeks) are marked as a small sample and not ranked: below that the pooled rate of the listing ages weighs more than a row's own clicks (the weight fitted on their own clicks).

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

Figure 8

Who may apply: location rules

Can people apply to roles in Math & STEM from anywhere?

Location ruleOpen rolesShare of its open rolesAll AI training work
Open worldwide43
17.3%
18.5%
Worldwide except listed countries50
20.2%
18.1%
Only listed countries155
62.5%
63.0%
No rule stated0
0.0%
0.4%

Sample. 248 roles in Math & STEM open at the end of 31 Oct 2026, against 4,381 across all AI training work.

The rule is read from each posting's location text, then, where that names no country, from what the posting's own description states, and where both are silent from what the platform states about where it hires.

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

Figure 9

Who may apply: roles open to each country

How many roles in Math & STEM can a resident of each country apply to?

CountryOpen rolesShare of its open roles
Any country (worldwide roles only)
43
17.3%
United States
145
58.5%
United Kingdom
112
45.2%
India
112
45.2%
Germany
104
41.9%
Philippines
100
40.3%
Indonesia
98
39.5%
Pakistan
90
36.3%
Argentina
76
30.6%
Colombia
75
30.2%
Chile
73
29.4%
Brazil
71
28.6%
Egypt
71
28.6%
Japan
70
28.2%
South Korea
68
27.4%
Bangladesh
67
27.0%
Ghana
67
27.0%
Nigeria
67
27.0%
Saudi Arabia
67
27.0%
Cambodia
63
25.4%
Palestine
55
22.2%

Sample. 248 roles in Math & STEM open at the end of 31 Oct 2026, for residents of a fixed set of 20 countries.

A country's count is the roles open worldwide (including roles that state no rule, as on the job board), plus roles whose list names it, plus roles whose exclusion list does not.

The 20 countries are the same in every report of this series: those named most often by the location rules of open AI training roles on 9 Oct 2026, either as the only countries accepted or as excluded.

Median pay is over the roles open to that country that state an hourly rate (at least 3).

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

Figure 10

Expert fields asked for

Which professions and disciplines do roles in Math & STEM ask for?

Expert fieldOpen rolesShare of its open roles
Mathematics and statistics
93
37.5%
Physics and astronomy
40
16.1%
Data science and machine learning
11
4.4%
Software and IT
11
4.4%
Engineering (non-software)
9
3.6%
Biology and life sciences
6
2.4%
Chemistry
5
2.0%
Finance and accounting
5
2.0%
Teaching
4
1.6%
Earth and environmental science
3
1.2%
Business and management
2
0.8%
Nuclear, CBRN and defense
2
0.8%
Humanities and social sciences
1
0.4%
Psychology and mental health
1
0.4%

Sample. 172 of the 248 roles in Math & STEM open on 31 Oct 2026 name an expert field in their title (69.4%); 14 expert fields are named.

A role counts for a field when its title names the profession or discipline; generic titles ('Expert', 'AI Tutor') name none. A role naming two fields counts for both.

Median pay needs at least 3 roles stating an hourly rate.

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

Figure 11

What the descriptions ask for

Do roles in Math & STEM ask for degrees, credentials, experience or tests?

RequirementRolesShare of described roles
Mentions a PhD147
63.4%
Mentions a PhD or master's177
76.3%
Mentions any degree187
80.6%
Mentions a professional credential10
4.3%
States years of experience20
8.6%
Asks for 5 years or more2
0.9%
States weekly hours90
38.8%
Mentions an assessment or test65
28.0%
Mentions an AI-led or video interview9
3.9%
Mentions a background check1
0.4%

Sample. 232 roles in Math & STEM open on 31 Oct 2026 whose job board publishes a full description (AfterQuery, Aligned Labs, Anthropic, Anyone AI, Braintrust, Claru, DataAnnotation, Handshake AI, Invisible, Mercor, micro1, Mistral AI, OpenTrain AI, Prolific, RemoExperts, Rise Data Labs, Scale AI Human Frontier Collective, Sovrano AI, SuperAnnotate (SME), Surge AI, TELUS Digital AI, Terac, TransPerfect DataForce, Turing, Vetto, Weekday and xAI (SpaceXAI)), 93.5% of its open roles; 4,000 described roles across all AI training work.

A mention can be a requirement or a preference ('a PhD is a plus'): the figure counts both. The patterns are those of the requirements report.

Only some platforms publish full descriptions, so this figure describes their roles, not every platform's.

A kind of work's figure is shown when its described roles outweigh the average of the kinds (5.8 described roles: the weight of every kind's shares around the shares of all described roles, fitted on their own counts).

Source. AITraining.jobs Research, The Market for Math & STEM Work, October 2026 edition.

For labs, data vendors and researchers

Buy the data behind this report

The figures above are a summary. Behind them are records we collect ourselves, every day:

  • 11,482 roles tracked, each with its history

    Every role on 62 job boards, read since 13 Jul 2026 (every hour since 18 Oct 2026): its pay and place, and the first and last day it was open.

  • 58,632 apply clicks in 28 days

    Which roles people actually apply to, by field, pay and kind of work. It happens on this site, so nobody else can count it.

  • 39 lab and vendor hiring boards

    The job boards of 25 AI labs and 14 human-data vendors, read every day: who is hiring for human data, and who is hiring to buy it.

What you can buy

  • The records

    The rows behind any figure, with their history, not only the totals.

  • A regular feed

    The same figures every week or month, as files.

  • A custom cut or benchmark

    Your countries, languages or fields, or your own platform against the market.

Building AI and need the experts behind this data? (you contract them yourself) or (we deliver the expert work).

Investing in AI-data companies? Hiring radar is these records role by role, every day, as files.

Method and definitions

The public job boards of the AI training platforms this site tracks, read every hour. Apply clicks are made on AITraining.jobs; full descriptions come from the platforms whose boards publish them. Every figure is computed by one database query over the window shown, the same query for the live report and for every edition. Each figure states its sample beneath it.

Kind of work
Each role is classified from its title and description into one of the twelve kinds of AI training work this site tracks; this report counts the roles classified as Math & STEM.
Fixed panel
The platforms already tracked when the weekly series opens. The index follows their open roles, so a platform added later (such as OpenTrain AI and Welo Data in September 2026) never reads as growth.
Index
The panel's average open roles in a week, divided by the same in the base week, times 100; shown beside the same index for all AI training work on the same platforms.
Open role
A role listed by a given day and still in its platform's feed that day.
New listing
A role that first appeared during the window on a platform already tracked before that week. A platform's first week in this dataset is its back catalogue and is not counted as new. When this site changes how it reads a platform's board (a cap lifted, a page kind or department read too, a relevance rule changed), the roles that came into view or left it from that change to the end of its week are this site's reading, not the platform's hiring: they are not new listings or removals, and changes in open roles are measured like for like across it. Each platform's report says what changed.
Removed listing
A role that left its platform's feed during the window.
Days listed
From a removed role's listing date to the last day it was seen in its feed. A median needs at least 3 removed roles.
HHI
Herfindahl-Hirschman Index: the sum of every platform's squared percentage share of open roles, 0 to 10,000. Under 1,500 unconcentrated, 1,500 to 2,500 moderately, above 2,500 highly concentrated.
Advertised pay
The hourly figure a posting states (the middle of a stated range), counted once per distinct rate as the job board counts it. The median and typical pay are the board's: typical pay runs from the 25th percentile of what roles pay (the middle of each advertised range) to the 90th-percentile advertised ceiling, stated from 3 distinct rates.
Clicks per open role a week
Apply clicks on AITraining.jobs on a specific role, divided by the days the roles were open, times 7.
Demand index
Clicks per open role a week divided by the same rate across every role on the site. Above 1, the roles draw more applicants than the average role.
Listing age
When a role was listed relative to the window's first day: during the window, up to 4 weeks before it, 4 to 12 weeks before or longer. The listing date is the job board's posting date, or else the first day the role appeared in the feed.
Location rule
Read from each posting's location text and, where that names no country, from what the posting's own description states: open worldwide, worldwide except listed countries, only listed countries, or no rule stated.
Open to a country
A role is open to residents of a country if it is open worldwide, its list names the country, or its exclusion list does not. The 20 countries compared are the same in every report of the series.
Window
Live: the last 28 complete days, in UTC, and the last twelve complete ISO weeks (from the week of 3 Aug 2026 at the earliest) for the weekly series. Edition: one calendar month, in UTC, and its weeks.

What this data cannot tell you

  • Platforms that recruit without a public job board are not counted. Several large ones invite workers to projects directly.
  • A listing is not a hire. Some platforms keep evergreen roles open for months, and some post one listing per language or country.
  • Apply clicks measure interest among the people who use AITraining.jobs, and what is shown first gets clicked more. Clicks are never published by platform.
  • Full descriptions are published by a few platforms only, so the requirements figure describes their roles and appears only when this kind's described roles read on their own against the other kinds' (the weight of every kind's shares around the shares of all described roles, fitted on their own counts).
  • A kind of work with few open roles gives figures that move with a handful of postings: rows on too little data are marked as a small sample.

How the job feeds are read and roles are classified: site methodology. How fresh, complete and consistent the data is, checked every day: data quality and coverage.

Editions

The live report covers the last 28 complete days and changes every day. Each complete month is also published as an edition, frozen two days after the month ends, so a figure you cite from an edition never moves.

Cite and reuse

Free to quote, chart and republish under CC BY 4.0 with a link to this page. This edition is frozen, so its figures will read the same later.

Citation

AITraining.jobs (2026). The Market for Math & STEM Work, October 2026 edition. https://aitraining.jobs/research/math-stem-ai-training-market/2026-10

BibTeX

@misc{aitrainingjobs_math_stem_ai_training_market_2026_10,
  author = {{AITraining.jobs}},
  title = {The Market for Math & STEM Work, October 2026 edition},
  year = {2026},
  howpublished = {\url{https://aitraining.jobs/research/math-stem-ai-training-market/2026-10}},
}
Data: every figure as CSV (one line per value) and JSON, sent to your work email.Pay and open-roles data

More research

The same figures for every other kind of work:

All 81 research reports