Do tech internships ask for AI skills? 1 in 4 do
How many tech internship and new grad postings mention LLMs, AI agents, or AI coding tools, by field, and how to show those skills on an application.
By Kevin Tang · Updated November 2026
The short answer
About a quarter of them mention AI skills now. Of 2,363 open tech internship descriptions we read on November 6, 2026, 24 percent mentioned LLMs, generative AI, AI agents, prompt engineering, or AI coding tools such as GitHub Copilot or Cursor, and so did 25 percent of 1,333 new grad descriptions. LLMs and generative AI came up most, in 17 percent of internships, then AI agents or retrieval-augmented generation (14 percent) and AI coding tools (6 percent). The field matters: about 30 percent of software and AI, ML, and data internships mentioned these skills, against 9 percent of hardware internships, and new grad software roles mentioned them most, at 37 percent. Most roles do not need AI research experience, but a project that uses an LLM API, and comfort with AI coding tools, now help in a quarter of applications.
AI skills have moved into entry-level tech job postings quickly. We read thousands of open tech descriptions to see how often they ask for AI skills now, which ones, and in which fields.
What we measured
On November 6, 2026, we read the full description of every open U.S. posting on the SimplifyJobs internship and new grad trackers, the feed behind Arial's job board, that applies through Greenhouse, Ashby, Lever, or Workday, one per role, leaving out contract AI-training gigs: 2,363 internships at 772 companies and 1,333 new grad roles at 590. A skill counts when the description names it anywhere, as a requirement or a plus. We skipped sentences about the hiring process, such as rules on using AI in interviews, and sentences describing the company itself, and did not count Microsoft's office Copilot as a coding tool, so only the work itself counts. (SimplifyJobs internship tracker, November 6, 2026 file, and the job pages of its open postings on Greenhouse, Ashby, Lever, and Workday, read that day; SimplifyJobs new grad tracker, November 6, 2026 file, and the job pages of its open postings on Greenhouse, Ashby, Lever, and Workday, read that day)
Which AI skills tech postings mention
Scroll sideways to see every column
| The description mentions | Internships | New grad roles |
|---|---|---|
| LLMs or generative AI | 16.5% | 14.8% |
| AI agents or retrieval-augmented generation | 13.8% | 13.8% |
| PyTorch, TensorFlow, or JAX | 9.1% | 5.7% |
| AI coding tools, such as GitHub Copilot or Cursor | 6.2% | 8.5% |
| Fine-tuning or training models | 5.5% | 4.1% |
| Prompt engineering | 3.9% | 2.9% |
| Any of LLMs, agents, prompting, or AI coding tools | 24.2% | 24.5% |
Postings that mention AI mostly ask about building with models, not training them: mentions of LLMs and AI agents far outnumber mentions of fine-tuning, and in new grad postings AI coding tools now appear more often than PyTorch. A typical line, from a Palo Alto Networks new grad posting: "Exposure to large language models (LLMs) or AI/ML APIs, even through personal projects." (SimplifyJobs internship tracker, November 6, 2026 file, and the job pages of its open postings on Greenhouse, Ashby, Lever, and Workday, read that day; SimplifyJobs new grad tracker, November 6, 2026 file, and the job pages of its open postings on Greenhouse, Ashby, Lever, and Workday, read that day)
AI skills by field
Scroll sideways to see every column
| Field | Internships mentioning AI skills | New grad roles mentioning AI skills |
|---|---|---|
| Software | 30.8% | 37.3% |
| AI, ML, and data | 29.1% | 29.4% |
| Product | 19.8% | 31.4% |
| Quant | 16.7% | 7.0% |
| Hardware | 8.6% | 6.3% |
Software roles lead, and new grad software roles most of all: 37 percent mentioned these skills and 16 percent named AI coding tools. AI, ML, and data roles mention LLMs most, about a fifth to a quarter of them. Hardware and quant postings mostly do not ask yet. The internship employers that mentioned AI skills in the most postings were Micron, NVIDIA, Visa, General Motors, and Toyota Research Institute. (SimplifyJobs internship tracker, November 6, 2026 file, and the job pages of its open postings on Greenhouse, Ashby, Lever, and Workday, read that day; SimplifyJobs new grad tracker, November 6, 2026 file, and the job pages of its open postings on Greenhouse, Ashby, Lever, and Workday, read that day)
Read nextEvery skill tech postings ask for, by field
How to show AI skills on an application
- Build one small project that calls an LLM API, such as a tool with retrieval or an agent, and put it on your resume with what it does.
- Name the AI coding tools you actually use, if the posting mentions them.
- Use the posting's own words: "LLMs", "generative AI", "agents", or "RAG", so both people and screening tools see the match.
- Do not claim research you have not done; a real project you can explain beats a long list of buzzwords.
Arial applies for you: on Greenhouse, Ashby, Workday, Oracle, and Indeed postings, the agent fills in the employer's form from your resume and profile, writing open answers from what your resume shows, asks you only what it cannot answer, and submits it. If a role asks for AI skills, the AI work on your resume is what those answers draw on.
AI skills now show up in a quarter of tech postings, mostly as building with models rather than researching them. One real project and honest familiarity with the tools covers what a quarter of postings now ask for.