# Capital One: AI workforce and talent

Source: https://www.bankingnewsai.com/banks/capital-one/ai-workforce
Last updated: Oct 31, 2026

*Part of the [Capital One AI strategy profile](https://www.bankingnewsai.com/banks/capital-one).*

## How is Capital One building its AI workforce?

Capital One trains most of its AI talent in-house and on a large base: it had approximately 76,300 employees at December 31, 2025, up from 51,987 two years earlier after the May 2025 Discover acquisition, and a technology workforce of more than 14,000, mostly engineers. In 2026 it opened an enterprise-wide AI learning hub to more than 60,000 associates and ran what it called its 'most intensive engineering learning initiative to date', a month-long agentic coding program that brought more than 10,000 engineers into one live session, with hands-on training on frontier models through partnerships with Google and Anthropic. The groundwork is older: Tech College, launched by CIO Rob Alexander in 2017 with machine learning among 13 disciplines, and a 160-hour Machine Learning Engineering Training Program whose first cohort of 40 retrained software and data engineers in 2022. On hiring, Evident counted more than 2,200 professionals added to Capital One's AI staff in its 2025 report, largely through Discover, and ranked its AI talent pool second among banks after JPMorgan Chase. The job cuts on the record are integration cuts, not AI ones: at least 1,139 Discover employees in 302 job titles, announced February 23, 2026. Capital One has not published an AI headcount, training hours or any estimate of jobs changed by AI. ([Capital One via SEC EDGAR](https://www.sec.gov/Archives/edgar/data/927628/000092762826000024/cof-20251231.htm), [SEC EDGAR](https://www.sec.gov/Archives/edgar/data/927628/000092762824000094/cof-20231231.htm), [Capital One](https://www.capitalone.com/tech/serverless/), [Capital One](https://www.capitalone.com/tech/ai/advancing-ai-fluency), [Capital One](https://www.capitalone.com/tech/software-engineering/from-the-cios-view-building-a-nimble-learning-organization/), [Capital One](https://www.capitalone.com/tech/machine-learning/accelerating-machine-learning-engineering-with-a-new-capital-one-skills-program/), [Banking Dive](https://www.bankingdive.com/news/banking-AI-talent-pool-recruitment-evident-bofa-capital-one-citi-jpmorgan-wells/802383/), [Banking Dive](https://www.bankingdive.com/news/capital-one-discover-job-cuts-layoffs/813908/))

## Which AI and machine learning training programs does Capital One run?

Programs Capital One has described in its own posts, oldest first. Scale figures are the bank's.

| Program | What it is | Scale | Since | Sources |
| --- | --- | --- | --- | --- |
| Tech College | Engineer-designed curriculum in 13 disciplines including machine learning/AI, cloud, cybersecurity and data; online courses, workshops, boot camps and partner training from Udacity, General Assembly and Coursera; 70% of the deans are engineers | Rolled out to thousands of tech associates in 2017, to be open to all 50,000; nearly 11,000 learners and thousands of classes by 2021; hundreds of courses in 2026 | 2017 (curriculum begun 2016) | [Capital One](https://www.capitalone.com/tech/software-engineering/from-the-cios-view-building-a-nimble-learning-organization/), [Capital One](https://www.capitalone.com/tech/machine-learning/the-machines-are-learning/), [Capital One](https://www.capitalone.com/tech/ai/advancing-ai-fluency) |
| Machine learning nanodegree pilot | Six months at 10 hours of study a week, inside Tech College | 300 learners registered at launch | 2017 | [Capital One](https://www.capitalone.com/tech/software-engineering/from-the-cios-view-building-a-nimble-learning-organization/) |
| ML 101 and the practitioner path | An introductory video series covering the ML life cycle and responsible use, internal case studies from Capital One data teams, and a hands-on learning path for engineers fluent in Python and GitHub | Not disclosed | Described May 2021 | [Capital One](https://www.capitalone.com/tech/machine-learning/the-machines-are-learning/) |
| Machine Learning Engineering Training Program (MLETP) | 160 hours, instructor-led and self-paced, on ML algorithms, MLOps, industry tools and Capital One platforms; a Coursera certificate and four projects; part of formalizing the machine learning engineer job family, with a path to open ML roles | First cohort of 40; third cohort fall 2022 | 2022 | [Capital One](https://www.capitalone.com/tech/machine-learning/accelerating-machine-learning-engineering-with-a-new-capital-one-skills-program/) |
| Enterprise AI learning hub | From foundational AI literacy to advanced workflows and domain-specific tools; learning built into routines 'backed by dedicated time' | More than 60,000 associates | 2026 | [Capital One](https://www.capitalone.com/tech/ai/advancing-ai-fluency) |
| Agentic coding program | Month-long: one live session, then weekly dedicated time, workshops and coding challenges against real enterprise problems; expanding to the agentic developer experience, multi-agent orchestration and token efficiency | More than 10,000 engineers | 2026 | [Capital One](https://www.capitalone.com/tech/ai/advancing-ai-fluency) |
| Frontier-lab training partnerships | Hands-on training with frontier models, tools and workflows for technologists | Google and Anthropic named as partners | Described August 2026 | [Capital One](https://www.capitalone.com/tech/ai/advancing-ai-fluency) |

## What has Capital One disclosed about its workforce and AI talent?

Company figures are from the 10-K human capital section and Capital One's own posts; AI-talent figures are Evident's, as reported.

| Measure | Value | As of | Source | Sources |
| --- | --- | --- | --- | --- |
| Associates (employees) | About 50,000 | September 2017 | Rob Alexander, Capital One Tech | [Capital One](https://www.capitalone.com/tech/software-engineering/from-the-cios-view-building-a-nimble-learning-organization/) |
| Technology organization | 11,000 people, 85% engineers | December 2020 | AWS; SiliconANGLE | [AWS](https://aws.amazon.com/solutions/case-studies/capital-one-all-in-on-aws/), [SiliconANGLE](https://siliconangle.com/2020/12/01/capital-one-closes-all-data-centers-to-rely-on-aws-on-demand-infrastructure-reinvent/) |
| Employees worldwide | 51,987 | December 31, 2023 | 2023 10-K | [SEC EDGAR](https://www.sec.gov/Archives/edgar/data/927628/000092762824000094/cof-20231231.htm) |
| Employees worldwide | Approximately 76,300 (includes Discover, acquired May 2025) | December 31, 2025 | 2025 10-K | [Capital One via SEC EDGAR](https://www.sec.gov/Archives/edgar/data/927628/000092762826000024/cof-20251231.htm), [Capital One](https://www.capitalone.com/about/newsroom/capital-one-completes-acquisition-of-discover/) |
| Technology workforce | More than 14,000, mostly engineers | 2025 | Banking Dive; Capital One | [Banking Dive](https://www.bankingdive.com/news/capital-one-software-databolt-data-security-platform/756909/), [Capital One](https://www.capitalone.com/tech/serverless/) |
| AI staff added | More than 2,200, largely through Discover; second-largest AI talent pool among 50 banks | October 2025 | Evident, via Banking Dive | [Banking Dive](https://www.bankingdive.com/news/banking-AI-talent-pool-recruitment-evident-bofa-capital-one-citi-jpmorgan-wells/802383/) |
| AI index rank (talent is one of four pillars) | Second of 50 banks for the fourth year | October 6, 2026 | Evident, via American Banker | [American Banker](https://www.americanbanker.com/news/jpmorganchase-still-tops-ai-ranking-as-big-bank-lead-widens) |
| Discover integration cuts | At least 1,139 employees in 302 job titles, announced February 23, 2026, including vice presidents of finance, technology, enterprise risk and enterprise services; 1,748 Illinois layoffs scheduled between October 2025 and October 2026 | March 5, 2026 | Banking Dive (Illinois WARN filings) | [Banking Dive](https://www.bankingdive.com/news/capital-one-discover-job-cuts-layoffs/813908/) |

## What has Capital One said about AI's effect on its jobs?

The bank's public line is about investment, not reduction. Fairbank told analysts in October 2025 that its AI opportunities 'require significant investment in AI and AI talent, and we are doing that', and the 2026 proxy lists 'talent and tools to deliver top tier software development, rigorous risk management, and artificial intelligence (AI) capabilities' among 2025's investments. ([The Motley Fool (Yahoo Finance)](https://finance.yahoo.com/news/capital-one-cof-q3-2025-225531123.html), [SEC EDGAR](https://www.sec.gov/Archives/edgar/data/927628/000119312526124131/d701179ddef14a.htm))

On how work changes, the bank says AI coding tools take over 'run-the-engine' tasks so engineers can focus on systems thinking, architecture and problem-solving, and that learning is designed not to compete with delivery. Natarajan said in September 2026 that customer-service agents now edit 'just one-fifth' of the copy AI drafts for them. ([Capital One](https://www.capitalone.com/tech/ai/advancing-ai-fluency), [Fortune](http://fortune.com/2026/09/29/how-banks-are-using-ai-agents-copilots-roi))

Not disclosed: an AI headcount, AI training completion or hours, or any role reduction attributed to AI. The 2025 10-K human capital section describes recruiting, peer-to-peer learning and quarterly attrition reviews without mentioning AI; the 2026 cuts on the record are tied to the Discover integration, and Capital One said it gave at least 60 days' notice and enhanced severance. ([Capital One via SEC EDGAR](https://www.sec.gov/Archives/edgar/data/927628/000092762826000024/cof-20251231.htm), [Banking Dive](https://www.bankingdive.com/news/capital-one-discover-job-cuts-layoffs/813908/))

## What does the Capital One profile record on this topic?

| Date | Move | Detail | Sources |
| --- | --- | --- | --- |
| Sep 29, 2026 | Natarajan: cutting latency and token costs with smaller, open-weight models | Chief enterprise AI officer Prem Natarajan says the bank has put more emphasis on agentic AI infrastructure to reduce latency and lower token costs, building customised open-weight and specialised models with less context targeted at the desired output; customer-service agents now edit only one-fifth of AI-drafted copy. | [Fortune](http://fortune.com/2026/09/29/how-banks-are-using-ai-agents-copilots-roi) |
| Aug 3, 2026 | AI fluency program | 'We don't just use AI—we build AI'; sustaining an edge requires more than off-the-shelf models. | [Capital One](https://www.capitalone.com/tech/ai/advancing-ai-fluency) |
| Nov 20, 2025 | Applied AI research agenda published | LLM training, customisation and alignment; synthetic data; multi-agent reasoning. | [Capital One](https://www.capitalone.com/tech/ai/applied-ai-research) |
| Oct 21, 2025 | Q3 2025 call: AI 'deeply embedded', not bolted on | Fairbank: AI-driven experiences are 'a new front' in premium cards; the opportunities require 'significant investment in AI and AI talent'. | [The Motley Fool (Yahoo Finance)](https://finance.yahoo.com/news/capital-one-cof-q3-2025-225531123.html) |
| May 18, 2025 | Discover acquisition completes | Adds more than 2,200 AI professionals to the talent pool, per Evident. | [Capital One](https://www.capitalone.com/about/newsroom/capital-one-completes-acquisition-of-discover/), [Banking Dive](https://www.bankingdive.com/news/banking-AI-talent-pool-recruitment-evident-bofa-capital-one-citi-jpmorgan-wells/802383/) |
| Mar 15, 2024 | $3 million Columbia center for responsible AI in finance (CAIRFI) | PhD fellowships, faculty research and public symposia; kicked off March 28, 2024. | [Capital One](https://www.capitalone.com/tech/ai/responsible-ai-partnership-columbia-university/), [Columbia Engineering](https://engineering.columbia.edu/about/news/kicking-center-responsible-ai-and-finance) |
| Nov 1, 2023 | Natarajan at the Senate AI Insight Forum | Four pillars; AI risk managed through model risk management, compliance, fair-lending and privacy governance. | [Capital One](https://www.capitalone.com/tech/ai/us-senate-ai-insight-forum/) |
| Dec 1, 2020 | Last data center closed; all in on AWS | Eight data centers exited after an eight-year migration; an 11,000-person technology organisation, 85% engineers. | [SiliconANGLE](https://siliconangle.com/2020/12/01/capital-one-closes-all-data-centers-to-rely-on-aws-on-demand-infrastructure-reinvent/), [AWS](https://aws.amazon.com/solutions/case-studies/capital-one-all-in-on-aws/) |
| Sep 26, 2018 | Why Eno's language technology was built in-house | The bank says it built Eno's natural-language processing itself, from concept to launch in under three months, training it on customer-service chat logs; Eno understood more than 2,200 terms and emojis for a balance request. | [Capital One](https://www.capitalone.com/tech/machine-learning/capital-ones-intelligent-assistant-why-we-built-enos-nlp-tech-in-house/) |

## Common questions

### How does Capital One train employees in AI?

Through Tech College, its internal learning program since 2017, and, from 2026, an enterprise-wide AI learning hub open to more than 60,000 associates. Engineers went through a month-long agentic coding program that started with a live session for more than 10,000 of them, followed by weekly dedicated learning time. Capital One also partners with Google and Anthropic for hands-on training on frontier models, and since 2022 has run a 160-hour program that retrains software and data engineers as machine learning engineers.

### How many employees does Capital One have?

Approximately 76,300 worldwide at December 31, 2025, per the 2025 10-K, up from 51,987 at the end of 2023; the increase reflects the Discover acquisition completed in May 2025. Its technology workforce is more than 14,000, mostly engineers, up from 11,000 when it closed its last data center in 2020.

### Is Capital One cutting jobs because of AI?

No cut has been attributed to AI. The 2026 reductions on the record are Discover integration cuts: at least 1,139 employees in 302 job titles announced February 23, 2026, and 1,748 Illinois layoffs scheduled from October 2025 to October 2026, per Illinois filings reported by Banking Dive. Capital One has not published any estimate of roles changed or removed by AI.

### How big is Capital One's AI talent pool?

Capital One does not publish one. Evident's 2025 report said it added more than 2,200 professionals to its AI staff, largely through Discover, and ranked its AI talent pool second among 50 large banks, behind JPMorgan Chase; Evident's October 2026 index again ranked Capital One second overall. Chief scientist Prem Natarajan joined from Amazon in 2023.

### Is Capital One hiring for AI roles?

Yes. BankingNewsAI's weekly read of Capital One's careers site on October 27, 2026 listed director-level AI engineering, generative AI data science, agentic AI product management, AI Foundations research and AI risk management roles; the current list is on the Capital One profile and the bank AI hiring tracker. Postings change weekly, so they are not used as sources here.
## Deeper on Capital One

- [How does Capital One use AI in auto finance?](https://www.bankingnewsai.com/banks/capital-one/auto-finance-ai)
- [What is Capital One's approach to responsible AI?](https://www.bankingnewsai.com/banks/capital-one/responsible-ai)
- [What AI research, patents and partnerships does Capital One have?](https://www.bankingnewsai.com/banks/capital-one/ai-research-patents-partnerships)
- [What is Capital One's virtual assistant and how does it compare with other banks'?](https://www.bankingnewsai.com/banks/capital-one/virtual-assistant)
- [What is Capital One's technology stack?](https://www.bankingnewsai.com/banks/capital-one/technology-stack)
- [What proprietary AI solutions has Capital One built?](https://www.bankingnewsai.com/banks/capital-one/proprietary-ai)
- [What is Capital One's AI strategy and who sets it?](https://www.bankingnewsai.com/banks/capital-one/ai-strategy)


---

Canonical page: https://www.bankingnewsai.com/banks/capital-one/ai-workforce
Part of [BankingNewsAI](https://www.bankingnewsai.com/) — a free daily brief on AI in banking, an AI regulation tracker (45 authorities, 260 documents) and AI-strategy profiles of the 120 largest US banks. Markdown versions of every reference page: append `.md` to the page URL; index at https://www.bankingnewsai.com/llms.txt.