Privacy-friendly analytics help us see which jobs people explore. They stay off unless you agree.Analytics off unless you agree. Details

Read the record
Back to the river
Skip to sources
Time Machine

Computer science professors

Official title Computer Science Teachers, Postsecondary (SOC 25-1021)

Also called CS professor, CS lecturer, Computer science lecturer, CS faculty, Postsecondary CS instructor

In brief

In May 2025, BLS counted 35,480 US computer science professors, with median pay of $96,980 a year. Of the BLS counts since 2021, the highest was 37,600, in May 2021. The earliest figure here is an estimate of 805 from 1975, not measured the same way as today's count. The last full technology era for computer science professors: Zoom + remote learning + Gradescope, 2020 to 2023.

What the sources say

Official outlook

BLS projects employment to grow 4.9% from 2025 to 2035.

BLS Employment Projections 2025-35

AI exposure studies

Data as of BLS OEWS May 2025; BLS Employment Projections 2025-35. BLS publishes these counts as yearly survey snapshots and cautions against reading them as a precise trend.

Last updated .

Cite this page
APA

Future History. (2026, October 11). Computer science professors: jobs and pay since 1975 [Data set; data as of BLS OEWS May 2025; BLS Employment Projections 2025-35]. https://futurehistory.earth/timeline/25-1021

Chicago

Future History. "Computer science professors: jobs and pay since 1975." Data as of BLS OEWS May 2025; BLS Employment Projections 2025-35. Last modified October 11, 2026. https://futurehistory.earth/timeline/25-1021.

BibTeX

@misc{futurehistory_25_1021, author = {{Future History}}, title = {Computer science professors: jobs and pay since 1975}, year = {2026}, month = oct, howpublished = {\url{https://futurehistory.earth/timeline/25-1021}}, note = {Last updated 2026-10-11. Data as of BLS OEWS May 2025; BLS Employment Projections 2025-35}, }

The access date is today's date on your device.

drag to travel through time
19752000now
2026
Known then as Mathematician / Numerical Analyst teaching computing (pre-department era)
Known then as Computer Science Faculty (early department era, often joint appointments with Math or EE)
Known then as Computer Science Professor / Lecturer (discipline maturation era)
Known then as Computer Science Teachers, Postsecondary (BLS SOC 25-1021)
Known today as Computer Science Teachers, Postsecondary (BLS SOC 25-1021)
Projected to still be called Computer Science Teachers, Postsecondary (BLS SOC 25-1021)
Data from the May 2025 survey
US Employment
35K
Est.

Stated in source: BLS OEWS

Official statistic
Report an error
Median Annual Wage
$96,980
Est.

≈ $94,494 in 2024 prices (derived by us, CPI-adjusted)

Official statistic
Report an error
Beat · 1953

Cambridge University launches the Diploma in Computer Science, the world's first full-year taught computing course, built around the EDSAC stored-program machine. The teaching team included David Wheeler and was led by faculty associated with Maurice Wilkes. There were no textbooks and no precedent; the course was literally invented alongside the machine. This is conventionally treated as the origin point of the CS teaching profession.

Beat · 1962

On October 24, 1962, Purdue University's Board of Trustees approves the first Department of Computer Sciences in the United States, with five founding faculty members: Samuel Conte, Richard Kenyon, L. Duane Pyle, Robert Korfhage, and Saul Rosen. An initial class of 24 master's and doctoral students enrolls. The curriculum is partly invented in real time. By 1963 additional faculty are hired; by the early 1970s the department has 15 full-time faculty, a size that remained representative of leading CS departments through the late 1970s.

Beat · 1965

Sister Mary Kenneth Keller earns a PhD in Computer Science from the University of Wisconsin-Madison in June 1965, becoming the first woman to earn a CS PhD in the United States. Karl Usow at Purdue earns the first CS PhD from that institution in June 1966. In 1979 the entire US and Canadian university system produces only 248 CS PhDs per year, fewer than 100 of whom choose academic careers, against a backdrop of over 650 unfilled academic positions. The PhD pipeline would remain the binding constraint on CS department growth for the following 40 years.

Beat · 1979

A National Science Foundation analysis documents the "seven-positions-for-every-applicant" crisis in CS faculty hiring: over 650 academic positions open nationwide, fewer than 100 CS PhDs choosing academic careers. Undergraduate enrollment in CS is growing at 15-20% annually. Departments respond by hiring faculty retrained from mathematics and electrical engineering, and by relying heavily on graduate teaching assistants. Nearly 400 faculty members from other disciplines participate in NSF-funded CS retraining programs in summer 1984 alone.

Beat · 1984

CS undergraduate enrollment reaches its first peak then begins a decline. Bachelor's degrees in CS peak around 1984 and fall 42% by 1994, a pattern that would repeat in 2004 (post dot-com bust) and has so far not repeated in the 2010-2024 boom. Departments that expanded rapidly in 1982-1984 find themselves overstaffed by the late 1980s; some lecturers hired during the boom are not renewed. The boom-bust cycle in CS enrollment becomes a defining structural feature of the discipline, with faculty headcount as the lagging indicator.

Beat · 2000

The BLS introduces SOC 25-1021 as a separately tracked occupational code for Computer Science Teachers, Postsecondary under the 2000 Standard Occupational Classification system. For the first time, it becomes possible to track national employment and wages for this occupation consistently across all employer types. The year 2000 anchor of approximately 31,800 employed reflects the peak of the second enrollment boom, just before the dot-com bust. By 2004-2005, employment in this code had modestly contracted as departments trimmed adjunct and visiting positions in response to lower student demand.

Beat · 2012

The third CS enrollment boom begins, driven by tech-industry salary premiums and cultural attention on coding (Code.org launches, "learn to code" becomes a mainstream career narrative). Coursera and edX launch MOOCs with initial claims that they will displace traditional CS instruction at scale; by 2015-2016 the completion-rate data (typically below 10%) ends the displacement thesis. The lasting effect of MOOCs is as a supplemental resource and as a competitive pressure that accelerated the quality of free online CS content. The third enrollment boom brings CS to approximately 10% of bachelor's degrees at some universities by 2020.

Beat · 2023

GitHub Copilot and ChatGPT arrive in CS classrooms simultaneously in spring 2023 with no preparation time. The SIGCSE 2025 Technical Symposium concludes that the assessment-redesign challenge is the central professional responsibility for CS faculty in the current era: existing assignment formats (write a function from scratch, implement a classic algorithm) are fully solvable by LLMs, and the pedagogical question of when students should encounter AI tools relative to foundational skill acquisition is unresolved. The ACM ICER 2025 brownfield study finds students complete tasks 34.9% faster with Copilot but do not understand the code they submit, identifying the core risk.

How to read this number

OEWS is a point-in-time survey snapshot, not a continuous time series; BLS advises against using it for year-over-year trend comparison.

Each dot is a cited figure over time; the dotted line only links them (values between aren't measured). Hollow dots are estimates.

Official statistic: Copied by a script from a statistics agency table or a published study's table, not retyped by hand. BLS figures and the forecast and exposure tables are also re-checked against that table on every data run. Researched context: Researched from the cited source rather than copied from it by a script, so it is not machine-checked yet. Spot a mistake? Each figure has a “Report an error” link.

Cited forecasts · present → 2035

Will AI replace computer science professors? What the research says

Each figure is stated by the named source; we publish no forecasts of our own. Where sources disagree, we show the disagreement rather than an average.

Each figure is stated by the named source; we publish no forecasts of our own. Where sources disagree, the spread is the honest signal.

What published forecasts say
Forecast change in the number of people doing this work, as published by each named source. Bars share one scale on every role, centred on zero and reaching the end at ±30%. Grey marks a small change, 3% or less either way.
BLS Employment Projections 2025-35
→ 2035
+4.9%
BLS National Employment Matrix, 2025-35 cycle, line for SOC 25-1021 (Computer science teachers, postsecondary): 43.9 thousand jobs in 2025, projected 46.1 thousand in 2035, a change of +2.2 thousand (+4.9%). Matrix employment counts all jobs in the occupation, including self-employed workers (0.2% of this occupation in 2025). The average for all occupations in this cycle is +3.5%.
Official statisticReport an error
How exposed is this work to AI (research measures)
Scores as published by each study. Each row says what it measures. These measure what AI could technically do, not forecasts of jobs lost.
Eloundou et al. (2023), "GPTs are GPTs"
2023 study · Share of tasks exposed to LLMs (β, human-rated)
36.9%
of tasks
What this measures

Published occupation-level exposure from the paper's occ_level.csv for O*NET-SOC 25-1021.00 Computer Science Teachers, Postsecondary. Shown: human-rated β = 0.369, the share of the occupation's tasks an LLM could do in at least half the time at equal quality, counting tasks that need LLM-powered software at half weight (β = E1 + 0.5 × E2). Other published measures for this occupation: human-rated α (LLM alone) 0.143 and γ (all LLM-assisted) 0.595; GPT-4-rated β 0.512. This measures how much of the work an LLM could touch, not how many jobs disappear.

Official statisticReport an error
Today, in this role

What's changing day to day for computer science professors

Three parts of your work where AI is already doing real lifting, and what stays yours.

AI is sitting alongside you hereGrade and provide feedback on large programming assignment submissions using Gradescope autograder pipelines, writing and maintaining test harnesses, configuring rubrics, reviewing AI-suggested answer groupings, and calibrating the manual grading that follows automated scoring.

Grade and provide feedback on large programming assignment submissions using Gradescope autograder pipelines, writing and maintaining test harnesses, configuring rubrics, reviewing AI-suggested answer groupings, and calibrating the manual grading that follows automated scoring.[11],[12]

Tools picking this up
Where your edge is

Shift grading effort from mechanical execution to rubric authorship and edge-case review: the autograder runs the test harness, but the faculty member must design the test cases that actually measure learning rather than output compliance. Instructors save 30-50% of grading time: reinvest it in harder feedback on design decisions.

AI is sitting alongside you hereHold office hours and asynchronous Q&A (via LMS discussion boards or Ed Discussion) to debug students' conceptual misunderstandings, triaging which questions AI tutors already answered adequately and reserving human engagement for the genuinely stuck student who has tried AI-assisted debugging and still cannot resolve their confusion.

Hold office hours and asynchronous Q&A (via LMS discussion boards or Ed Discussion) to debug students' conceptual misunderstandings, triaging which questions AI tutors already answered adequately and reserving human engagement for the genuinely stuck student who has tried AI-assisted debugging and still cannot resolve their confusion.[5],[13]

Tools picking this up
Where your edge is

Redirect office-hour time toward the harder class of problems: students who are stuck because they have a deep conceptual gap, not because they cannot find the right syntax. AI tutors are excellent at the latter; only a human expert can identify and address the former. Track which question types still escalate despite AI availability: those are your highest-leverage teaching moments.

Get started with these tools
AI is sitting alongside you hereConduct and publish original computer science research (formulating novel research questions, implementing experiments, analyzing results, and writing papers) using AI tools (ChatGPT Edu, NotebookLM) to accelerate literature synthesis and drafting without outsourcing the intellectual contribution.

Conduct and publish original computer science research (formulating novel research questions, implementing experiments, analyzing results, and writing papers) using AI tools (ChatGPT Edu, NotebookLM) to accelerate literature synthesis and drafting without outsourcing the intellectual contribution.[6],[14]

Where your edge is

Use AI for literature mapping (NotebookLM can summarize many papers into a structured synthesis in minutes) and for drafting related-work sections; keep the research question formulation, experimental design, and interpretive contribution as the human core: reviewers and tenure committees assess novel intellectual contribution, not productivity.

Where computer science professors go next

Natural next steps for someone with your foundation: not exits, evolutions.

What each move takes

A direction you could grow

Education Administrators, Postsecondary

CS faculty frequently move into department chair, associate dean, or dean of engineering/computing roles, especially as institutions urgently need administrators who understand AI's impact on curriculum and can make credible technology decisions. The pivot is natural for faculty who enjoy service work and governance, and who want to shape institutional policy on AI adoption, AI ethics curriculum, and accreditation standards. CS faculty bring rare technical credibility to administrative roles that are increasingly concerned with AI governance.

What you'd add
  • · Higher education budget management and resource allocation
  • · ABET accreditation self-study and continuous improvement processes
  • · Faculty performance review and promotion/tenure process facilitation
  • · Strategic enrollment management and curriculum market positioning
  • · University AI governance policy development and vendor evaluation
What it takesSome new skills to pick up
Tools of the era

How has the work of computer science professors changed since 1953?

The tools that defined the work, one line per era. Select an era to see how it reshaped the work.

  • Batch mainframe + punched cards (teaching on the machine itself)

    The earliest CS instructors taught by example on the machine: EDSAC at Cambridge, IBM 701 and 704 at US universities, and later the IBM 7090. There was no textbook in the modern sense. Students submitted programs on punched cards, waited hours or days for batch processing, and received output on paper printouts. The instructor's role was inseparable from the machine operator's: explaining why a program failed often required tracing the physical execution of the machine. Course content was defined by whatever the available hardware could do, which meant heavy emphasis on numerical computation, machine-language coding, and assembly programming.

    Effect on the work

    The batch-turnaround model limited how many students a single instructor could effectively teach: classes were small because debugging sessions required individual attention and machine time was scarce and expensive. This kept the ratio of faculty to students high and the total number of CS positions small.

  • Timesharing terminals + UNIX (interactive computing enters the classroom)

    The ARPA-funded Compatible Time-Sharing System at MIT (1961) and the later spread of UNIX (Bell Labs, 1969; licensed to universities from 1975) transformed CS teaching from deferred-batch to interactive. Students could now log into a timesharing system, see their program execute immediately, and iterate in minutes rather than days. UNIX provided a common pedagogical environment that standardized the C language and systems programming curriculum across US universities. The DEC PDP-11 and VAX became the canonical teaching machines of the era: Purdue, for instance, acquired its first VAX 11/780 in 1978. Interactive computing allowed class sizes to grow moderately but required faculty who could teach live debugging, the shell, and file systems.

  • Personal computers in labs + Turbo Pascal / C (democratized lab computing)

    The IBM PC (1981) and Apple II made it possible to build dedicated computing labs where every student had their own workstation, eliminating the timesharing bottleneck and making CS courses accessible to non-science majors. Curriculum shifted toward structured programming languages (Pascal, then C) that could run on personal hardware. The personal-computer era coincided with the first undergraduate enrollment boom: entering CS freshman enrollment peaked at Purdue in fall 1981 at hundreds of students, creating pressure that forced mass lecture formats and the use of teaching assistants as the primary student contact for most undergraduates.

    Effect on the work

    Lab computing reduced the marginal cost of each additional student, enabling class sizes to grow substantially. Faculty headcount grew much more slowly than enrollment, with TAs and lecturers absorbing most of the surge. This structural pattern, high student-to-tenure-line-faculty ratios bridged by graduate TAs and adjuncts, defined CS departments through the following four decades.

  • Internet + web-accessible development tools (the dot-com teaching era)

    The World Wide Web brought CS into undergraduate general education and drove a second enrollment surge from 1996-2001. Instructors added HTML, networking, and web-application development to curricula that had been predominantly systems and algorithms. Online course management systems (Blackboard, WebCT, then Moodle) changed how problem sets were distributed and collected. For the first time, homework submission, grade publication, and course announcements moved fully online. The bust brought enrollment back down by 2003-2004 and briefly stabilized the faculty market.

    Effect on the work

    The dot-com boom and bust is the clearest demonstration that CS faculty employment is a lagging indicator of industry demand: department expansion decisions take several years to execute (faculty search, PhD pipeline, building space) while enrollment responds in a few years, creating persistent boom-bust cycles in both headcount and curriculum.

  • Open-source ecosystems + MOOCs (Python, GitHub, and the third enrollment boom)

    Python's rise to the default introductory language by the early 2010s (replacing Java, which had replaced C++, which had replaced C, which had replaced Pascal) was the most significant pedagogical infrastructure shift of the era. Python's readability lowered the on-ramp for introductory courses and opened CS to non-majors in data science, biology, and economics. GitHub normalized version-control as part of intro courses. The lasting MOOC effect was as a supplemental resource and as proof that CS content could travel at scale without replacing the faculty relationship. Meanwhile the third enrollment boom, driven by the tech industry's decade of rapid hiring, brought CS to a small share of all bachelor's degrees in some university systems by 2020.

    Effect on the work

    OEWS faculty headcount stayed roughly flat through this period (33,080 in 2010, 32,230 in 2020) and stood at 36,240 in 2024. New hiring was dominated by lecturer and adjunct positions rather than tenure-track lines: research shows that between 2000 and 2016 most new postsecondary teaching positions were contingent (non-tenure-track), a trend that ran across all disciplines including CS.

  • Zoom + remote learning + Gradescope (COVID-forced digital pedagogy)

    COVID-19 forced the entire postsecondary system fully online in March 2020, exposing the structural gaps in digital pedagogy that CS departments had never needed to solve because their labs were physical. Live coding demonstrations migrated to shared-screen sessions in Zoom; office hours became async chat threads; assessment integrity collapsed as students submitted code from home with unlimited resource access. Gradescope, which had been growing in CS departments since 2015, became essential: its autograder pipeline ran test harnesses against submitted code at scale and returned results without requiring synchronous sessions. Proctoring software (ProctorU, Honorlock) was briefly adopted but generated significant student resistance and documented false-positive plagiarism flags.

    Effect on the work

    The COVID period increased the technical burden on faculty substantially: maintaining Zoom sessions, reconfiguring autograders, managing academic integrity in an all-remote environment, and building asynchronous course materials added several hours per week to a typical course load by multiple faculty surveys. It also accelerated the adoption of Gradescope, which subsequently became standard in CS departments well after in-person instruction returned.

  • Large language model coding assistants (GitHub Copilot, ChatGPT Edu, Codio Coach)

    GitHub Copilot, released publicly in June 2022, and ChatGPT, launched November 2022, arrived simultaneously in classrooms in spring 2023, with no preparation time for curriculum redesign. An ACM ICER 2025 study found that students using GitHub Copilot on brownfield code tasks completed assignments 34.9% faster but reported not understanding why the AI-generated suggestions worked. The SIGCSE 2025 consensus was clear: AI tools enrich learning only when faculty redesign the assessment loop, and that redesign is a genuinely novel professional task requiring CS expertise and real-time pedagogical judgment. ChatGPT Edu, deployed system-wide by California State University to 460,000 users in February 2025, and Indiana University to 120,000 users, represents the first institution-scale adoption of a specific AI tool as infrastructure for both teaching and research.

    Effect on the work

    The introduction of LLM coding assistants has not reduced faculty demand; it has increased the complexity and novelty of the faculty role. Assessment redesign, academic integrity adjudication, and curriculum sequencing (when do students first encounter AI tools relative to foundational skill acquisition?) are new high-stakes professional responsibilities that did not exist before 2023.

Part of Education · see all 51 roles →
Different role?

See the same long-arc view for your own profession.

Browse the directory by industry, or search by title or SOC code.

Browse all roles

Computer Science Teachers, Postsecondary employment and pay by year

On record since1953
Latest US employment35,480 (May 2025)Source: BLS OEWS May 2025Official statisticReport an error
Latest US median pay$96,980 (May 2025)Source: BLS OEWS May 2025Official statisticReport an error
Outlook+4.9% by 2035Source: BLS Employment Projections 2025-35Official statisticReport an error
View all 28 years of cited US data, each linked to its source
YearUS employmentAnnual pay (median unless marked)SourceReport
1975805n/aEst., Stanford CS Capacity Project Researched contextReport
1979837n/aEst., Stanford CS Capacity Project Researched contextReport
1980n/a$28,000Est., AAUP Annual Report on the Economic… Researched contextReport
1990n/a$52,000Est., ERIC EJ408775 Researched contextReport
200027,770$68,000BLS OEWS 2000 (employment) Official statisticEst., Approximate 2000 median annual salary… (pay) Researched contextReport
200334,470$51,030BLS OEWS May 2003 Official statisticReport
200437,260$52,630BLS OEWS May 2004 Official statisticReport
200538,520$54,270BLS OEWS May 2005 Official statisticReport
200636,630$57,620BLS OEWS May 2006 Official statisticReport
200733,840$62,020BLS OEWS May 2007 Official statisticReport
200832,520$66,440BLS OEWS May 2008 Official statisticReport
200932,240$68,580BLS OEWS May 2009 Official statisticReport
201033,080$70,300BLS OEWS May 2010 Official statisticReport
201133,510$73,090BLS OEWS May 2011 Official statisticReport
201234,350$72,200BLS OEWS May 2012 Official statisticReport
201335,770$73,100BLS OEWS May 2013 Official statisticReport
201435,410$72,010BLS OEWS May 2014 Official statisticReport
201533,760$74,840BLS OEWS May 2015 Official statisticReport
201632,540$77,570BLS OEWS May 2016 Official statisticReport
201732,230$78,630BLS OEWS May 2017 Official statisticReport
201832,430$82,220BLS OEWS May 2018 Official statisticReport
201931,800$85,180BLS OEWS May 2019 Official statisticReport
202032,230$85,540BLS OEWS May 2020 Official statisticReport
202137,600$77,910BLS OEWS May 2021 Official statisticReport
202233,870$84,760BLS OEWS May 2022 Official statisticReport
202336,150$96,430BLS OEWS May 2023 Official statisticReport
202436,240$96,690BLS OEWS May 2024 Official statisticReport
202535,480$96,980BLS OEWS May 2025 Official statisticReport
Embed this timeline on your site

Free for any site. Paste this where the timeline should appear; it stays interactive, every datapoint stays cited, and it sets no cookies on your page. How embedding works

<iframe src="https://futurehistory.earth/embed/25-1021"
  width="100%" height="430" style="border:0"
  title="Computer Science Teachers, Postsecondary, a Future History timeline"
  loading="lazy"></iframe>
<p style="font-size:13px;margin:4px 0 0"><a href="https://futurehistory.earth/timeline/25-1021">Computer Science Teachers, Postsecondary: jobs and pay over time</a>, from Future History</p>

See all roles in Education