AI Career Observatory
AI Labor Market Report

State of AI Careers 2026

The overall shape of the AI job market: role distribution, skill demand, remote share and salary disclosure, from tracked job postings.

Generated Sep 29, 2026 · Dataset window: Dec 5, 2009 – Sep 29, 2026 · Observations: 3,302

Executive summary

Analysis of 3,302 active AI job postings from 15 companies, collected from public employer career pages and ATS systems. AI Agent Engineering and Forward Deployed roles are the fastest-moving new categories in the tracked dataset.

Role distribution

The tracked market splits across 12+ canonical roles. The table below shows the largest role categories by active job count.

RoleActive jobsShare
Forward Deployed Engineer2317.0%
AI Solutions Engineer2156.5%
AI Go-To-Market Specialist682.1%
AI Research Engineer601.8%
Applied AI Engineer341.0%
AI Engineer310.9%
Customer Engineer300.9%
Data Engineer260.8%
AI Agent Engineer230.7%
Data Scientist190.6%
Technical Account Manager180.5%
Machine Learning Engineer170.5%

Skills demand

Skill frequency is extracted from full job descriptions using pattern matching. The leaders are consistent with an application-layer market: Python, LLMs and RAG lead, with agent tooling (MCP, LangGraph, LangChain) already visible.

SkillJobs mentioningShare of all jobs
Python87526.5%
Evaluation72522.0%
LLM65219.7%
AWS58317.7%
AI Agents56017.0%
Google Cloud50715.4%
Azure49815.1%
SQL48214.6%
System Design47314.3%
Spark35010.6%
Java2748.3%
Kubernetes2607.9%
Observability2447.4%
TypeScript2417.3%
Fine-Tuning1795.4%

Remote work

460 of 3,302 tracked jobs (13.9%) are marked remote by the employer. Hybrid arrangements dominate for on-site-listed roles at frontier labs.

Salary disclosure

563 postings (17.1%) disclose a concrete salary range. Disclosure is concentrated in US-based employers, consistent with state pay-transparency laws.

Methodology & limitations

Dataset: active job postings collected from public employer career pages and ATS APIs (Greenhouse, Lever). Role classification uses exact-title pattern matching with evidence stored per posting; skills are extracted from description text.

Limitations: the dataset covers tracked employers only and is a lower bound on the market. Figures describe the tracked dataset, not the entire AI labor market. Salary figures use advertised ranges only.

Full details: data methodology.