CronJobs original data · AI/ML hiring pulse

AI/ML software engineering jobs: September 2026

CronJobs observed 117 eligible U.S. AI/ML software engineering jobs on September 1, 2026—7.1% of its 1,655-job inventory. This is a descriptive snapshot, not evidence that AI caused software-engineering employment to rise or fall.

Observed September 1, 2026 at 19:42 UTC · Methodology version 2026-09-01 · By the CronJobs engineering team · Editorial policy

What did the snapshot find?

117eligible AI/ML roles
7.1%of eligible inventory
4thlargest of seven role categories
1,655eligible jobs in the denominator

AI/ML was the fourth-largest normalized role category, behind backend (847), DevOps/SRE (345), and security (183). The categories are mutually exclusive in this role-mix table; the technology tags below are not.

Which AI-related technology tags appeared?

These are raw normalized tag counts from eligible job descriptions. A single job may contribute to several rows, so the counts must not be added together. The pipeline also keeps distinct normalized labels such as llm and llms separate in this snapshot.

Published technology tagEligible jobs containing tag
machine learning76
llm71
llms53
pytorch49
rag36
tensorflow30
ai/ml26
ai21

Is AI reducing software engineering jobs?

This snapshot cannot answer that question. It records one eligible inventory at one time. Measuring impact requires comparable observations over time and a design that separates AI effects from hiring cycles, interest rates, industry composition, and other causes.

Current external evidence also needs careful scope. The U.S. Bureau of Labor Statistics projects software-developer employment to grow 15.8% from 2024 to 2034 and says AI adoption is expected to support demand in some computer occupations; a projection is not a count of current openings or proof of a single cause. Read the July 2026 BLS projection.

Indeed Hiring Lab reported that U.S. software-development postings increased almost 15% from late February 2025 while the overall posting index declined, with senior roles and AI-titled roles accounting for much of the increase. Indeed also described the rebound as coming from a low base and warned that correlation does not establish causation. Read the July 2026 Indeed analysis.

What does this page not measure?

  • It does not measure employment, hires, layoffs, applications, offer rates, or jobs outside the CronJobs source inventory.
  • It does not combine technology-tag synonyms or infer that a technology mention is a primary job requirement.
  • It does not compare periods, so it cannot establish growth, decline, seasonality, or the impact of AI.
  • It does not cross-tab AI/ML roles by seniority, work mode, salary, company, or source because those aggregates are not in the published snapshot.

Data, method, and citation

The figures are a focused reading of the immutable September 2026 U.S. software engineering jobs report. Review the jobs data methodology, source registry, role-mix CSV, and technology-demand CSV before reusing them.

Suggested attribution: CronJobs AI/ML Software Engineering Jobs snapshot, observed 2026-09-01T19:42:13.310Z, derived from the September 1, 2026 CronJobs U.S. Software Engineering Jobs dataset, methodology version 2026-09-01.

Browse AI/ML Jobs in Discord

AI/ML software engineering jobs FAQ

How many AI and machine learning software engineering jobs were in the September 2026 CronJobs snapshot?

At September 1, 2026 at 19:42 UTC, CronJobs contained 117 eligible U.S. AI/ML software engineering jobs. They represented 7.1% of the 1,655 eligible jobs in that point-in-time inventory.

Which AI and machine learning technology tags appeared most often?

Among the published raw normalized tags, “machine learning” appeared in 76 eligible jobs, “llm” in 71, “llms” in 53, PyTorch in 49, RAG in 36, and TensorFlow in 30. Counts overlap because one job can contain multiple tags, and spelling variants are not combined.

Does this snapshot prove that AI is creating or eliminating software engineering jobs?

No. One point-in-time inventory cannot measure change over time or establish causality. It shows that AI/ML roles and AI-related technology requirements were present in the eligible CronJobs inventory on the observation date.

How were the AI/ML job and technology counts produced?

The counts use the same conservative U.S. software-engineering eligibility, seven-day source-freshness, role-classification, duplicate-handling, and ten-job publication rules documented in the CronJobs jobs data methodology.