Every hiring manager I speak to in market infrastructure tells me some version of the same thing.

"We've been trying to fill this role for six months."

Then I ask how many applicants they received.

"About 200."

Here's the problem. That number is almost meaningless.

The illusion of supply

When a company posts a Market Data Engineer role, or anything touching exchange connectivity, feed handlers, low-latency infrastructure, the application pipeline fills up fast.

200 CVs. Sometimes more.

On paper, it looks like a functioning talent market.

It isn't.

Because when you actually filter that pipeline for genuine domain capability, here's roughly what you find:

  • 150 have never touched trading data in their lives
  • 30 have some capital markets exposure, but it's peripheral
  • 10 understand data pipelines but have no idea what an order book actually does
  • 5 have real domain experience
  • 1 or 2 might actually be the right hire

That's not a talent shortage. That's talent dilution.

Hundreds of engineers look relevant. A handful actually are.

Why the curve is so steep

Market infrastructure engineering isn't software engineering with a trading skin on top.

It requires a completely different mental model. One built around:

  • Feed handlers and exchange protocols
  • Order book construction and tick normalisation
  • Latency constraints measured in microseconds
  • Market microstructure and how venues actually behave
  • Vendor architectures at firms like Bloomberg, LSEG, ICE Data Services, FactSet

Engineers who have worked inside these environments aren't learning the domain. They're operating inside it.

That difference sounds subtle. In practice, it's the difference between a 12 to 18 month ramp-up and someone who can contribute from week one.

Three things making this worse

This dynamic isn't static. The curve is getting steeper.

1. The systems are more complex than they used to be

Modern market data infrastructure involves cloud distribution, multi-venue ingestion, normalised data layers, real-time analytics, and historical tick replay, often in the same architecture. The depth of knowledge required has increased significantly. So has the gap between generalist and specialist.

2. The market is expanding

Fixed income electronification. ETF liquidity infrastructure. Consolidated tape initiatives in Europe. Alternative data vendors. Derivatives data. Every one of these creates demand for the same small pool of domain engineers.

3. The best people rarely move through job adverts

The most experienced market data engineers tend to sit deep inside Bloomberg, LSEG, ICE, and their ecosystem firms. They're well compensated. They're embedded. When they move, and they do move, it happens through networks, referrals, and targeted conversations.

Not job boards.

The misdiagnosis that costs companies months

When hiring stalls, the default conclusion is usually: we need to go more senior.

So companies raise the seniority bar, repost the role, and get another 200 applications with the same underlying problem.

Seniority doesn't fix this.

What actually matters is domain proximity, whether someone has worked directly inside exchange feeds, market data distribution, or trading system infrastructure.

A mid-level engineer who has spent three years at a market data vendor will routinely outperform a senior engineer moving in from general fintech. The ramp-up time alone makes the seniority premium irrelevant.

What this means in practice

The Market Data Scarcity Curve explains why standard hiring methods consistently fail in this niche.

Open pipelines create the illusion of supply. Keyword filtering makes it worse, it surfaces engineers who have touched adjacent technologies without any real domain depth.

The talent that matters for these roles is concentrated in a small, specific part of the market. It moves through networks. It doesn't respond to generic outreach.

Which means the hiring approaches that work here look different:

  • Targeted market mapping rather than reactive posting
  • Direct outreach to people with traceable domain history
  • Networks built inside the specific ecosystem, not just "fintech"
  • Positioning that signals genuine domain knowledge, not just technical requirements

The companies that hire well in this space understand that they're not fishing in a large pond. They're fishing in a very specific stretch of a very specific river.

The ones that don't understand that spend six months wondering why they keep catching the wrong thing.

Richard