Back to all posts

AI Job Postings Show Agents Moving Ahead of LLMs

AI appeared in more than half of November 2026 job postings, but the more revealing detail is which terms employers are choosing. Agent mentions outpaced LLM mentions—and that’s a shift in job-ad language, not proof that one technology has replaced another.

AI job postingsAI agentsLLMsMCPmachine learning
main thumbnail for AI Job Postings Show Agents Moving Ahead of LLMs
main thumbnail for AI Job Postings Show Agents Moving Ahead of LLMs
Reader Lens

Automation needs a narrow first win

The best first AI workflow is usually a repeated task with a clear input, clear output, and a human approval step.

AI job postings show how employers are describing AI-related work, not a direct count of AI systems being built or used. In the November 2026 snapshot, AI appeared in 51.6% of postings, compared with 30% two years earlier. The detail worth noticing is that mentions of AI agents now outpace mentions of large language models (LLMs)—but those are labels in job ads, not mutually exclusive kinds of work.

AI job postings now mention agents more than LLMs

AI agents, including postings that use the term “agentic,” appeared in 14.8% of November postings. LLMs appeared in 10.7%. The source article describes this as a shift in the work itself toward agents. That is a useful reading of the numbers, but the distinction needs care: a posting mentioning agents does not tell us exactly what the employee will build, and the figures do not show that agent work has replaced LLM work.

The gap does suggest that employers are naming agent-related work more often than they name LLMs in this snapshot. That matters because a job-searching reader might otherwise treat “AI” as one broad skill label. The postings point to more specific vocabulary being used, even though they cannot tell us how deeply each technology is used on the job.

inside paper visual for AI Job Postings Show Agents Moving Ahead of LLMs
main thumbnail for AI Job Postings Show Agents Moving Ahead of LLMs

Worth a look

Phugialy Picks

AI Engineering: Building Applications with Foundation Models
Amazon

AI Engineering: Building Applications with Foundation Models

A practical guide to building real-world applications with foundation models and LLMs.

GMKtec K15 AI Mini PC Oculink Intel Ultra 5 125U 32GB DDR5 512GB SSD | Desktop Computer AI Boost, 3X M.2 2280 Storage Expansion, Dual NIC...
Amazon

GMKtec K15 AI Mini PC Oculink Intel Ultra 5 125U 32GB DDR5 512GB SSD | Desktop Computer AI Boost, 3X M.2 2280 Storage Expansion, Dual NIC...

2084 and the AI Revolution, Updated and Expanded Edition: How Artificial Intelligence Informs Our Future
Amazon

2084 and the AI Revolution, Updated and Expanded Edition: How Artificial Intelligence Informs Our Future

Some Phugialy Picks use affiliate links. If you buy through one, Phugialy may earn a commission. It doesn't change what we recommend. Full disclosure →

MCP’s rise is striking, but tracking matters

MCP appeared in 4.5% of postings, up from 0.3% a year earlier. That is a sharp increase in the share of ads mentioning it. The source also says its tracking now includes terms such as RAG, LangChain, Anthropic, pgvector, and Vector Database, alongside AI Agents, LLM, and MCP.

Here’s the distinction that can get lost: a term being newly added to a tracker is not the same as the term being newly invented or newly adopted by employers. It means the analysis is now watching for it. The MCP comparison is especially notable because it includes figures from both the current snapshot and a year earlier; the list of newly tracked terms, by itself, is evidence about measurement coverage, not a trend for every item on that list.

Read the vocabulary shift, not a hiring forecast

The older “machine learning” label slipped from 14.8% of postings to 11.5%, while AI appeared in more than half. Taken together, those figures suggest a change in the language employers use to describe technical roles. They do not establish that overall AI hiring rose by the same amount, or that every posting mentioning AI requires the same capabilities.

My read is that the most useful signal here is not simply “AI is everywhere.” It is that the vocabulary is becoming more differentiated: agents, LLMs, and MCP are appearing as separate searchable terms, while the broader machine-learning label is less common. For candidates and people tracking the market, that makes exact wording worth paying attention to—but these percentages should be treated as clues about job-ad language, not a complete map of the work.

A useful next check is whether later snapshots keep showing agents ahead of LLMs and MCP above its earlier share. If those patterns persist, the vocabulary shift will look less like a one-month fluctuation and more like a durable change in how employers describe AI-related roles.

Holiday picks

Thanksgiving: early shopping

Start early: products for hosting, travel and home, plus the gear you have been watching.

We earn from qualifying purchases made through these links, at no extra cost to you, as an Amazon Associate and through other affiliate programs. Disclosure. See all Thanksgiving picks

Source and trust note

Built from source research and filtered through practical implementation judgment.

Reference: www.hntrends.com

Got a question about how this applies to you? →

Keep reading

Follow the thread