When Palantir embedded engineers inside military command centres, assigning them to integrate and provide support for the company’s data analytics platforms into classified workflows and mission-critical operations, it appeared like an expensive anomaly, a niche approach serving the most secretive customers on the planet.
But this model has now become the blueprint for AI companies to win enterprise deals worth millions. From OpenAI to Salesforce and MongoDB, companies are hiring Forward Deployed Engineers (FDEs), a role created initially by Palantir.
Some are even calling this the ‘hottest job’ in the generative AI space.
So what is the deal?
Simply put, FDEs are software engineers who embed themselves directly with customers to implement, customise, and deploy technical solutions within their specific business environments.
Unlike traditional engineers, FDEs work on-site or in close partnership with clients to bridge the gap between general-purpose technology and real-world operational needs.
For instance, Mistral, the French AI startup, is hiring for a forward-deployed machine learning engineer, who will participate in pre-sales calls to understand client needs and explain the company’s technology to various stakeholders.
Describing their experience, an FDE at Baseten, an AI infrastructure platform, said in a blog post, that when customers use their services, they often come up with ‘hairy technical challenges’ like ‘How can we optimise our model serving latency by 10x while keeping costs flat?’ or ‘What’s the best way to horizontally scale our generative AI pipeline to handle 100x more traffic?’
“After scoping the problem with the customer, we’ll work with them to design a solution and kick off the implementation,” they said.
In a conversation with AIM, Omkar Pandharkame, the chief strategy officer at Supervity AI, said that his company is aggressively hiring FDEs. “They are able to spin up demos very quickly, they are able to demonstrate the effectiveness of an AI agent…and win the confidence of the customer from a technical perspective,” he said, adding that the role is an evolution of a ‘pre-sales’ engineer.
Many in the industry have echoed a similar sentiment. Michelle Lim, a startup founder and an angel investor, said in a post on X that the FDE role isn’t new, but the evolution of the solutions engineer or a consultant. “But by rebranding it as FDE, AI startups have created an exciting career path for technical talent who want deep customer interaction,” she added.
Leo Mehr, director of engineering of Ramp, the financial operations platform, in a blog post said that before they hired FDEs, the scope of a customer request would be figured by an account executive, or a customer success manager, and then deliver it to the engineering team, who would then ‘scope out a mega-project’ that would take months to deliver. However, Mehr says that with FDEs, such issues have been prevented.
“A couple of months ago, one customer in onboarding was blocked by a feature gap that we estimated would take ~3 days of eng [engineering] work. We hopped on a call with them and figured out a workaround right there — if we had taken the request at face value, it would have needlessly cost us multiple engineer-days,” said Mehr.
Several companies are actively hiring FDEs, but each job description seeks a diverse set of requirements. Currently, there are over 200 job postings for FDEs on LinkedIn in the United States alone.
While the core requirements converge on software engineering skills, LLM integrations, and more, they differ in customer context. Some are public-sector with on-site work and clearances, and others are commercial SaaS with heavy prototyping and roadmap feedback.
But why FDEs now?
Enterprise customers are purchasing AI solutions, but often lack the expertise to implement them effectively. Legacy systems continue to pose a significant challenge, which explains the growing popularity of FDEs today.
“Even today’s sophisticated AI can’t autonomously navigate complex enterprise workflows and requirements and integrations,” said Lim in her post on X, adding that the ability to translate between business requirements and code has always been important.
Moreover, AI has also made building products and proof of concepts a relatively more straightforward process, meaning the real challenge is figuring out what customers and clients truly need. “The depth of your relationship with each customer matters more than ever,” noted Brianne Kimmel, an investor.
But the advantages flow in either direction. While FDEs can help customers make most of their money, they in turn help the company refine their products.
In an interaction with AIM, Ashish Kumar, chief data scientist at Indium, said that a crucial responsibility of an FDE is to capture “tribal knowledge” — the undocumented, highly specific ways an organisation runs critical workflows.
“On the face of it, something like invoice reconciliation looks standard,” Kumar explains. “But in reality, every company does it differently — where the invoice comes from, what APIs they use, what happens if there’s a failure, whether it falls back to Excel or a custom system. These are things you won’t find in documentation.” Kumar said that this tribal knowledge is used to improve the current agentic solution or the LLM systems.
According to him, “if you don’t capture this tribal knowledge from day one, especially for the edge cases, even a 98 or 99% accurate AI system will fail in production, because those edge cases are often the most critical.”
Kumar also said that FDEs, in their experience from interacting with customers, can recommend features to the product teams as well.
“FDEs need to be very good communicators, bring engineering, strategy and consulting skills at the same time,” he added.
The post What’s the Deal With the Forward Deployed Engineering Roles? appeared first on Analytics India Magazine.
