Wall Street’s AI Hiring Boom Has a New Prize: Engineers Who Coordinate Agents

Before AI takes over more banking work, it is creating a hiring challenge: finding people who can make teams of AI agents work together. References to that skill, known as agent orchestration, jumped 1,721% this year in bank job postings, according to an analysis by hiring-data company Draup provided exclusively to CNBC and reported by CNBC Economy (direct). The task is not simply building an AI tool. It is deciding how several specialized tools should share a job—and when a human must step in.

AI-related job postings at banks including JPMorgan Chase, Citigroup and Capital One rose 49% this year compared with 2025, reaching 139,819 listings, Draup found. The firm draws on public job advertisements and platforms including LinkedIn. Those figures measure advertised roles and references to skills, rather than completed hires. They point to banks moving beyond chatbots toward systems that can handle a larger share of everyday work.

From building models to making them useful

An earlier wave of recruitment centered on engineers and data scientists who built AI models or adapted them to company information. Now banks also want people who can put those systems to work inside individual businesses. These workers, often called forward-deployed engineers, combine technical knowledge with an understanding of a particular operation, whether a trading desk, a back office or a human-resources department.

Draup chief executive Vijay Swaminathan told CNBC that connecting agents inside a bank requires navigating complexity that is not always visible at first. One agent might inspect raw data, another review a document and another check compliance with regulations. Even automating employee vacation approvals can uncover numerous exceptions. The engineer must decide which agents are needed, assign their responsibilities, choose the technology and establish where human oversight belongs.

Coordinating AI agents ranks among Wall Street’s most sought-after skills, according to Draup CEO Vijay Swaminathan.

The demand extends to the tools behind these systems. References to LangGraph, a framework for building workflows with several steps, increased 679%. Mentions of LlamaIndex, which connects AI applications to data, rose 291%. Retrieval-augmented generation, or RAG—a method of supplying models with information from company databases—recorded a 259% increase. Swaminathan said banks also increasingly value problem-solving, creativity and employees willing to ask difficult questions about how a process actually works.

Controls become part of the hiring race

Banks are recruiting for safeguards as well as automation. References to responsible AI, meaning the safe and accountable use of these systems, climbed 657%, while mentions of AI governance and risk management rose 394% and 359%, respectively. Governance-related skills accounted for more than 16,000 references in Draup’s data, nearly twice the roughly 8,400 associated with training, deploying and running models. Swaminathan said cybersecurity concerns include vulnerabilities introduced by outside tools and connections to external models.

The pay reflects the competition for expertise. Roles involving generative AI—systems that produce content—and agents typically pay more than technology jobs elsewhere in finance, according to Draup. Generative AI managers receive a median base salary of about $190,000. Yet Swaminathan said these specialized positions remain difficult to fill, prompting major banks to rely heavily on retraining existing developers and employees with business expertise.

The next test is how that internal training translates into deployment and changes to existing jobs. JPMorgan CEO Jamie Dimon has discussed large-scale staff redeployment as AI takes on more work. For banks, the challenge now runs along two tracks: finding people who can coordinate agents and preparing employees for the work those systems leave behind.