Snowflake pays FDEs on three salary bands across two tracks

At Snowflake, the highest base salary for an Applied AI Forward Deployed Engineer is $181.7k a year. The lowest base salary for a Senior FDE is $200k.

The Applied AI role sits in the Cortex AI team in Menlo Park and builds production AI agents for Snowflake’s strategic customers. Yet the role’s entire salary band falls below the floor of the Senior role that migrates Spark workloads. At a glance, it is tempting to conclude that data pays better than AI.

That conclusion is hasty. Applied AI and Senior Spark differ in both seniority and track at the same time, so the gap between them says nothing on its own. The meaningful comparison lies elsewhere: between Senior Spark and Lead Migration, two roles on the same track.

If you are choosing an FDE track to invest the next few years in, learning to separate these two variables when reading job postings is worth more than any salary ranking.

Three bands that form a ladder

The aggregator aced.io (formerly Exponent’s FDE jobs board) lists three Snowflake FDE base salary bands as of September 2026. Applied AI FDE pays $126k–$181.7k, Senior pays $200k–$287.5k and Lead Migration FDE pays $236k–$339.25k.

The site does not say which track the Senior band belongs to. But the Senior FDE – Spark posting on Built In lists $200K–$288K, almost an exact match, so the Senior band can be read as the Spark role’s band. The Lead Migration posting on Wing’s jobs page lists $236k–$339,250 plus equity, matching aced.io.

Putting the figures together with the content of each posting, the three roles look like this:

	Applied AI FDE	Senior FDE (Spark)	Lead Migration FDE
Base salary	$126k–$181.7k	$200k–$287.5k (Built In: $200K–$288K)	$236k–$339.25k
Seniority word in title	None	Senior	Lead
Track	Applied AI (Cortex AI team)	Migration/Spark	Migration/Spark
Main work	Design, build and deploy enterprise AI solutions, including AI agents	Modernise customers’ Spark and data engineering workloads	Lead customer migrations to Snowflake, design scalable solutions
Way of working	Direct technical adviser to customers’ data science and engineering teams, at least 25% of time onsite	Unclear	Works with multiple FDE teams, each responsible for one customer

Only by combining the three postings does the structure become clear: the two roles on the right share a track and differ only in seniority. The gap between Applied AI and Senior mixes both variables.

Which step can be measured, and which cannot?

From Senior Spark to Lead Migration, the track stays the same and only seniority changes. The floor rises from $200k to $236k, an extra $36k; the ceiling rises from $287.5k to $339.25k, an extra $51.75k. This is the only comparison in the data set that separates seniority from track.

From Applied AI to Senior, both variables change at once. The floor differs by $74k and the ceiling by $105.8k. There is no way to know how much of that comes from the word “Senior” and how much from swapping AI agents for Spark workloads.

To know whether AI is paid less than data, you would need a Senior Applied AI band to set beside the Senior Spark band, and the public data does not currently include one. The $74k gap at the floor is more than double the Senior-to-Lead step, but these are two different career steps, so one cannot be subtracted from the other either.

Picture an engineer who writes good Python and has already put an agent system into real production. That person fits the Applied AI posting.

What does the migration track actually do?

A second Senior posting, tied to the SnowConvert AI tool, shows what the migration track leans towards. The work is described as designing and implementing data pipelines and modern, scalable data architectures on Snowflake. The tool carries the word AI, but the engineer’s job is still data engineering.

If this track really is paid more, one plausible hypothesis is risk. An AI agent that answers wrongly can often be fixed by adjusting the prompt, adding evals and running it again. A migration that goes wrong at the data model layer can skew reports, break downstream pipelines and cost the customer’s trust in the platform.

The Lead level also reveals the scale of responsibility: the posting describes the person working with multiple FDE teams, each focused on one customer.

If the risk hypothesis holds, the person leading migrations is responsible for the data of several customers at once. But that remains inference; the data proves only the seniority gap within the same track.

The number on the posting is not the number in the contract

Before taking these figures into a negotiation, keep three things in mind. Even the aggregators disagree: Built In lists the Senior Spark ceiling as $288K, aced.io as $287.5k. The difference is small, but it is a reminder that this data has passed through other hands.

Postings do not always include numbers, either. The SnowConvert AI posting gives no salary band and directs US readers to the Snowflake Careers site. Postings also vanish quickly: Built In removed the Senior Spark posting on 27 October 2026 and the Lead Migration posting on 30 October 2026, so save a copy while you can still see it.

And all of these are base salaries. Aced.io states clearly that equity comes on top and total compensation is not disclosed. Two people on the same base band can still have very different total pay, so the question about equity belongs in the first conversation with the recruiter.

Which track should you bet on?

If you already build LLM applications and want to join Snowflake, Applied AI is the posting with the lowest salary band, and also the only one of the three without the word Senior or Lead. In return, you must spend at least 25% of your time onsite as a direct adviser to customers’ data science and engineering teams. That is where customer discovery skills are built.

With no public Senior Applied AI band, it cannot yet be said that moving up to Senior requires switching tracks. Even so, an AI engineer who understands where a customer’s data lives and how it flows will have more options on both tracks.

When writing your CV, use the exact language of the posting you are targeting. For an Applied AI role, describe the agent you put into production and how you measured the results. For a migration role, describe the workload you moved: the data model before and after, how you reconciled the data, the rollback plan.

When reading any FDE posting, look at the word in front of “Forward Deployed Engineer” before the number after the dollar sign. Only when two postings share that word does comparing pay across tracks mean anything.