Hiring AI Talent for the Trading Desk: What Firms Need to Get Right
July 20267 min read
Hiring AI Talent for the Trading Desk: What Firms Need to Get Right
Hiring AI talent in sales and trading is becoming harder.
How do you define whether you need a machine learning engineer, an AI-literate trader, a data specialist embedded on the desk, or a team that can connect technical development with front-office priorities? And how do you find candidates with the right balance of technical depth, product knowledge and front-office judgment?
According to Rico Cottell, Principal Consultant – Data & AI Market Specialist at Selby Jennings, many financial institutions are still working through their strategy:
Everyone wants AI talent, but they do not always know why they want it or what they want that person to do. The hiring criteria can still be undefined, even when the technical bar is very high.
The wrong hire can leave firms with strong technical capability but little commercial value. Here’s how to ensure your next hire turns AI into better decisions, faster workflows and real trading impact.
Front-office workflow automation is driving AI hiring
Early financial services AI projects often focused on legal, HR, compliance and operational processes, but Rico is now seeing hiring move closer to trading desks and portfolio management teams:
The big push for the hedge funds and trading firms we work with is front-office workflow automation. They are looking at tasks from the start of a trade to the end and asking how they can streamline processes, improve efficiency, and accelerate execution.
This is creating demand for AI and machine learning talent that can work directly with salespeople, traders, structurers, researchers and portfolio managers. Rather than developing tools in isolation, these professionals need to understand how their work connects to decision-making, risk and revenue.
Example projects from Rico’s recent searches include:
- Building greenfield AI projects alongside fixed income and macro desks
- Developing machine learning models for portfolio managers
- Training and fine-tuning models for research and trading applications
- Automating market data collection, classification and distribution
- Applying generative AI and large language models to front-office workflows
- Testing computer vision, text and audio data as new research inputs
- Enhancing software and data infrastructure to enable AI adoption across trading teams
How AI hiring needs differ across FICC, equities and commodities
AI hiring requirements should be defined at an asset-class and desk level, as the data, market structure and commercial objective will directly inform the right profile.
AI talent in FICC
Fixed income, currencies and credit teams process large volumes of pricing data, economic releases, central bank communications, issuer information and market commentary.
AI hiring in FICC may support:
- FX, rates and credit research
- Fixed income and macro trading workflows
- Credit analysis and document review
- Repo, securities finance and financing solutions
- QIS and cross-asset structuring
- Market data automation
- Scenario analysis and model development
The required team could include machine learning engineers, data engineers and quantitative developers alongside FX, rates or credit professionals who can identify suitable use cases and assess model output.
AI talent in equities
Equity sales, trading and research teams can use AI to process company reports, earnings transcripts, news, research and alternative data.
Relevant use cases include:
- Supporting equity research and idea generation
- Extracting information from large document sets
- Improving equity trading and execution workflows
- Developing equity derivatives and QIS strategies
- Producing sales intelligence and client insights
- Supporting portfolio managers with research tools
- Automating market and reference data processes
Hiring AI talent in equities may require technical specialists embedded with cash equities, equity derivatives, research or portfolio management teams.
AI talent in commodities
As commodities businesses rely on complex datasets, AI talent needs to understand how physical market drivers affect pricing and risk. A technically strong candidate without power, oil, gas, metals or agricultural market knowledge may struggle to build tools that traders and commercial teams trust.
AI hires in commodities can support:
- Power and gas trading analytics
- Oil and refined products research
- Carbon market modelling
- Metals and agricultural commodity analysis
- Commodity origination and structuring
- Cross-commodity sales intelligence
- Trade finance and market data automation
Do finance firms need an AI engineer or AI skills within existing roles?
One of the first hiring decisions firms face is whether they need a dedicated AI position or stronger AI capability within an existing software, data, quant or front-office team.
Rico has seen banks take a different approach from some hedge funds and trading firms.
Rather than creating a standalone AI engineer vacancy, banks may ask existing software and data engineering teams to integrate new tools, build the required infrastructure and introduce AI into current workflows. Hedge funds and proprietary trading firms are more likely to build specialist AI engineering teams or place technical hires directly alongside investment teams.
The right structure depends on the objective:
| Hiring Need | Potential Profiles |
| Build and train proprietary models | AI engineers, machine learning engineers, AI researchers |
| Improve front-office workflows | Applied AI engineers, software engineers, data engineers |
| Support portfolio managers or traders | Machine learning engineers, quant developers, data scientists |
| Automate market data | Data engineers, market data specialists, software engineers |
| Connect technical and trading teams | Forward deployed engineers, technical product specialists |
| Apply AI within a specific market | AI-literate traders, researchers, salespeople and structurers |
Once firms have decided where AI capability should sit, the next challenge is defining the position precisely enough to target and assess the right candidates.
Why undefined AI job descriptions slow hiring in sales & trading
One of the most common hiring mistakes, Rico says, is starting a search with a broad request for a highly technical candidate, which only becomes more specific after the interview process starts.
“AI engineer,” “machine learning engineer” and “data scientist” can refer to very different levels of software engineering, modelling and production experience, so relying on job title alone creates several problems:
- Recruiters and internal talent teams target the wrong candidate pool
- Strong candidates receive conflicting information about the role
- Interviewers assess different skills at each stage
- Employers compare candidates with very different backgrounds
- The process takes longer while the business revises the specification
Before hiring AI talent in sales and trading, have the answer to the following five questions:
- What sales, trading or research problem will this hire address?
- Will they build models, infrastructure, tools or integrations?
- Which desk or market will use their work?
- What data and technology are already available?
- How will the business measure the hire’s impact?
This information will also help to decide how much financial markets experience the role requires.
The market is interested in junior AI talent, but the technical bar is high
Rico highlights growing competition for early-career AI professionals, as candidates closer to university research often have more recent exposure to generative AI, large language models and machine learning methods.
However, recent AI knowledge alone is not enough. Graduate talent may understand the latest methods but lack experience deploying systems in a production environment, while experienced software engineers may have stronger engineering fundamentals but less depth in machine learning. Rico explains further:
The sweet spot I am seeing is around one to three years of experience, with two to three years often working well. Firms want people close to the latest developments, but they still need strong computer science fundamentals.
For current compensation benchmarks, read our guide to AI talent salaries in financial services.
AI-assisted coding is making candidate assessment harder
AI coding assistants have increased productivity, but they have also made it harder to assess a candidate’s underlying technical ability.
Rico reports that some developers and engineers who use AI to write a large proportion of their code struggle when asked to complete technical tests independently:
AI-generated code is not always a true representation of someone’s knowledge. It may show how well they can prompt a tool, rather than how well they understand the code or the underlying problem.
To better evaluate machine learning and AI talent during the interview process, relevant discussions may include:
- The candidate’s personal contribution to previous projects
- Their understanding of software and computer science fundamentals
- How they selected, trained and evaluated a model
- Their ability to identify weak data or model failure
- Experience moving from a prototype to production
- How the work affected a trader, researcher or investment process
- Their ability to explain technical decisions to non-technical stakeholders
Infrastructure can determine if AI candidates accept a role
Hiring is always a two-way conversation, and strong AI candidates will want to understand the quality of the data, compute, software platforms and deployment processes, alongside access to market data, system ownership, front-office collaboration and the governance needed to move models into live workflows.
As Rico summarises:
If the infrastructure is not there, there is very little point unless the candidate is being hired to build it from scratch. You cannot do the job until the platform is ready.
Building a new AI platform can appeal to candidates who want ownership of a greenfield project, but they will want to see that the scope, budget, authority and internal support available has already been established.
Hire AI talent in sales and trading with Selby Jennings
Hiring AI talent is more than identifying strong engineers. It requires a deep understanding of where AI will create commercial value, the technical capabilities required to deliver it, and how to assess candidates in a rapidly evolving market.
Selby Jennings partners with banks, hedge funds, asset managers, proprietary trading firms and commodity businesses access specialist talent across trading, structuring, research, portfolio management, machine learning, data and AI engineering.
Our consultants can also support role design, talent mapping, candidate assessment, interview strategy and compensation benchmarking, helping firms define the right profile and hire people who can apply AI to real trading, research and client priorities.
Request a call back from Selby Jennings to discuss the AI talent your firm needs across your sales and trading desks.