New AI Job Titles · Pre-sales
Applied AI Architect
An Applied AI Architect is the technical advisor who helps an enterprise evaluate a frontier model and designs how it gets deployed, working from technical discovery through to a scalable architecture. Anthropic and OpenAI both hire the title. It is a pre-sales and advisory seat: the Architect designs the deployment, and a Forward Deployed Engineer embeds with the customer afterwards to build it.
What does an Applied AI Architect actually do?
The work runs from technical discovery to deployment design. An Architect sits with an enterprise that is evaluating whether a frontier model can do a specific job, establishes what the model can be trusted with, architects the proof of concept, and then designs a deployment that will still hold when the pilot becomes production. At Anthropic the role sits inside the Applied AI organisation and works across Sales, Product and Engineering, which tells you the seat is deliberately cross-functional rather than parked in one of them.
The distinguishing skill is knowing where a proof of concept stops scaling. Model capabilities demo well and productionise badly, and most of the value an Architect adds is catching that gap during evaluation instead of six months into a rollout. The AI Engineer owns the systems around the model; the Architect owns the decision about whether those systems should exist for this customer at all.
Why do Anthropic and OpenAI use the same title?
Because both companies hit the same organisational problem at roughly the same time, and because Solutions Architect was too generic to describe it. Anthropic has run several variants of the role, including Public Sector, Enterprise Tech, Industries and APAC postings, plus a Head of Applied AI Architects, Beneficial Deployments leadership seat. OpenAI’s San Francisco posting frames the same job differently: a senior technical owner for a named portfolio of enterprise customers, described internally as being the CTO of your book of business, guiding pre-sales through deployment and measured business impact.
Two competitors converging on one label is the part worth acting on. Where a single company coins a title, it is usually internal branding. Where two do it independently, the market has settled on a shape, and it is reasonable to expect the title to spread to the enterprises buying from them.
How is this different from a Forward Deployed Engineer?
| Applied AI Architect | Forward Deployed Engineer | |
|---|---|---|
| When | Before the deal | After the deal |
| Produces | Evaluation, proof of concept, deployment design | Working software |
| Sits | Across the customer’s stakeholders | Inside the customer’s team |
| Writes code in customer environment | No | Yes |
| Anthropic org | Applied AI | Applied AI |
Both sit under Applied AI at Anthropic, so the handoff happens inside one organisation rather than across a company boundary.
For a CTO on the buying side, that split is the useful read. The person who scoped your deployment and the person who will be in your Slack for the next quarter are different hires with different incentives, and it is worth knowing which one is in the room.
Should you hire this role yourself?
Only if you are on the selling side of an AI deployment, or you are large enough that your internal business units behave like customers. The role exists to bridge a commercial boundary. Inside a single company, the same work is usually the AI Transformation Architect, which owns reference architectures and platform standards rather than a customer portfolio. Naming an internal seat Applied AI Architect without the commercial boundary tends to produce an architect with no deals to architect for.
Role scope verified against live postings: Anthropic, Applied AI Architect · OpenAI, Applied AI Architect (San Francisco) (August 2026).
Applied AI Architect: common questions
Is Applied AI Architect just a Solutions Architect with a new name?
Largely yes, and the rename carries information. Solutions Architect is the generic title for a pre-sales technical advisor across all enterprise software. Applied AI Architect narrows the scope to foundation-model adoption: model evaluation, proof of concept, and a deployment design that survives contact with the customer’s data. The reason to notice the label is that two competing labs chose the same words independently.
How is an Applied AI Architect different from a Forward Deployed Engineer?
The Architect works before the deal and designs. The Forward Deployed Engineer works after it and builds, embedded in the customer team. Anthropic runs both roles inside the same Applied AI organisation, so the two are complementary rather than competing: the Architect produces the deployment design that the engineer then executes in the customer environment.
What background do these roles hire from?
The postings describe technical discovery through deployment with Sales, Product and Engineering, which is the working life of a senior solutions architect or a pre-sales engineering lead. What is added is frontier-model fluency: knowing what a model can be trusted to do, how to evaluate it against a customer’s task, and where a proof of concept quietly stops scaling.
Is there a leadership track for this title?
Anthropic has posted a Head of Applied AI Architects, Beneficial Deployments, so the function has at least one leadership seat above the individual-contributor line. That is a useful signal: companies do not create a head-of role for work they expect to fold back into an existing team.
What does an Applied AI Architect earn?
Compensation for this title is not yet publicly benchmarked the way AI Engineer or Software Engineer bands are. The sample is small, mostly under two years old, and concentrated at two companies. What can be said is that these are frontier-lab roles attached to enterprise revenue, which is not typically the low end of a pay structure. Treat any specific figure quoted for this title as unverified.
