What is Group Relative Policy Optimization (GRPO) and how common is it in AI job postings?

Group Relative Policy Optimization (GRPO) is a skill in the "ML Techniques" category. It currently appears in 28 active AI roles across 14 companies in our index.

Which companies are hiring for Group Relative Policy Optimization (GRPO)?

The top employers with active AI roles mentioning Group Relative Policy Optimization (GRPO) are: CrowdStrike (5), ByteDance (4), Scale AI (3), Amazon (3), Perplexity (2).

Is demand for Group Relative Policy Optimization (GRPO) growing?

Over the last 9 weeks, 20 new AI postings mentioned Group Relative Policy Optimization (GRPO). Demand is rising — up 100% in the last four weeks compared to the earliest four weeks in the window.

At which stage of AI development is Group Relative Policy Optimization (GRPO) used?

Roles requiring Group Relative Policy Optimization (GRPO) are concentrated in: agents (32%), post-training (29%), serving infrastructure (18%). These stages follow a seven-stage AI lifecycle from data preparation through to shipped product.

← All skills

Group Relative Policy Optimization (GRPO)

28 active AI roles across 14 companies mention Group Relative Policy Optimization (GRPO). Category: ML Techniques.

Top employers

CrowdStrike5Enterprise ByteDance4Big Tech Scale AI3Data AI Amazon3Big Tech Perplexity2AI Frontier NVIDIA2Semiconductors Adobe2Enterprise Together AI1Data AI Synthesia1Multimodal Salesforce1Enterprise Figma1Enterprise Deloitte1Consulting Apptronik1Robotics AMD1Semiconductors

Demand trend

New AI postings mentioning Group Relative Policy Optimization (GRPO) per week — 20 total over 9 weeks.

Sector

All Enterprise · 9 Big Tech · 8 Data AI · 4 Semiconductors · 3 AI Frontier · 2 Robotics · 1 Multimodal · 1 Consulting · 1

Function

All Engineering · 17 Research · 10 Product · 2

Status

All Active only

Sort

AI score Recently posted Company A–Z

FilteredfunctionResearch×

10 AI roles requiring this skill.

Company	Title	Sector	AI score	Stage
Together AI	Research Engineer, Core ML	Data AI	10	L3
Synthesia	Principal Research Engineer	Multimodal	9	L1
Scale AI	Research Scientist, Safety Post Training	Data AI	9	L2
Amazon	Applied Scientist - Agentic AI, Amazon Fulfillment Technology	Big Tech	9	L4
Scale AI	Research Scientist, AI Controls and Monitoring	Data AI	9	L5
Scale AI	Research Scientist, Agent Robustness	Data AI	9	L4
ByteDance	Research Engineer/Scientist (all levels), Efficient Models	Big Tech	9	L2
ByteDance	Sr. Research Engineer/Scientist (all levels), Efficient Models	Big Tech	9	L2
ByteDance	Sr. Research Engineer/Scientist (all levels), Efficient Models	Big Tech	9	L2
Figma	AI Applied Scientist	Enterprise	9	L2

Frequently asked questions

What is Group Relative Policy Optimization (GRPO) and how common is it in AI job postings?
Group Relative Policy Optimization (GRPO) is a skill in the "ML Techniques" category. It currently appears in 28 active AI roles across 14 companies in our index.
Which companies are hiring for Group Relative Policy Optimization (GRPO)?
The top employers with active AI roles mentioning Group Relative Policy Optimization (GRPO) are: CrowdStrike (5), ByteDance (4), Scale AI (3), Amazon (3), Perplexity (2).
Is demand for Group Relative Policy Optimization (GRPO) growing?
Over the last 9 weeks, 20 new AI postings mentioned Group Relative Policy Optimization (GRPO). Demand is rising — up 100% in the last four weeks compared to the earliest four weeks in the window.
At which stage of AI development is Group Relative Policy Optimization (GRPO) used?
Roles requiring Group Relative Policy Optimization (GRPO) are concentrated in: agents (32%), post-training (29%), serving infrastructure (18%). These stages follow a seven-stage AI lifecycle from data preparation through to shipped product.