Hello Reader,

One job title keeps coming up over and over from people trying to break into AI right now. Forward Deployed Engineer. It sounds impressive and vague at the same time, and most people applying for it do not actually know what it requires.

I sat down with Nacho, who has spent 7+ years in the AI industry and works closely with both candidates and companies hiring for these roles, to get a straight answer. Here is what actually matters if you want to become one.

Forget the job title, focus on the actual skill

The clearest definition : a forward deployed engineer goes into a company to solve a problem. That problem might require communication, implementation, or both, and it varies constantly. Given that range, the single most important skill is not a specific language or framework. It is knowing how to use AI efficiently across different types of problems.

That means having a mental map of what tools work for what situation. Know the integrations, libraries, and what connects to what. The question is not "which agentic framework should I master." It is "do I know which integrations are available for this specific problem."

The framework itself matters less than most people think. You will end up reusing 80 to 90 percent of the integrations you already know across future projects regardless of which framework wraps around them.

What actually gets you noticed for these roles

Build a portfolio that cuts against what most people assume. You do not need a groundbreaking, state-of-the-art project. You need something you keep improving over time that eventually gives you enough visibility for someone to say "I want to work with this person."

Nacho gave a real example from his own career: a small cybersecurity script, only 15 to 20 lines of JavaScript, built as a WhatsApp plugin to track chat history. It started with a simple demo, got posted on Reddit, someone made a TikTok about it, a YouTuber picked it up, and it snowballed into his most popular repository. The lesson is not "build something viral." It is "build something small, show the functionality clearly, and keep polishing it publicly." Momentum builds from consistent visibility, not from a single perfect launch.

The three layers of GenAI work, and where the real opportunity is

Nacho broke the GenAI space into three distinct layers, and this framing alone is worth internalizing if you are choosing where to focus.

Layer one is using AI. Connecting an application to a large language model so it can perform tasks. This is where most beginner projects live, and it is also the layer enterprise customers now consider too basic to hire for.

Layer two is engineering the agent harness itself. This is a level up: instead of building an application that uses an existing agent framework, you modify the framework or the harness directly. Improving the memory layer, improving context engineering, building a better agent loop, inducing skills from prior execution history. This requires more expertise, and it is specifically what enterprise customers are looking for because the first layer is already commoditized.

Layer three is applications that use AI tangentially to solve a bigger problem. RAG systems, semantic search, hybrid search, eval. These are real production use cases where AI is one component of a larger system rather than the entire point.

If you are trying to differentiate yourself, layer one is where most beginners get stuck. Layer two is where the real value and the real hiring demand currently sits.

You do not need a machine learning degree to do this work

Nacho's own path started in robotics and cybersecurity, not formal AI education. His advice for anyone with an existing technical background, especially cloud or infrastructure experience: you do not need to go all-in on machine learning theory. The application layer, using AI, building agents, understanding memory and context, is learnable through building and practice.

His most direct advice for someone already working in cloud or software: combine what you already know with GenAI rather than starting from zero. If you understand EC2, serverless, or Kubernetes, learn how to deploy an agent on that same infrastructure, how to call an MCP server from a Kubernetes pod, how cloud-native services integrate with agentic systems.

That combination, existing infrastructure knowledge plus applied GenAI, is what makes a candidate valuable, because no company adopts GenAI in a vacuum. It always sits on top of existing databases, applications, and architecture.

If you prefer watching or listening to the full conversation, here is the podcast:

​

Keep learning and keep rocking 🚀,

Raj

P.S - If you want to get an AWS Solutions Architect job without coding or learning every AWS service, the 10th cohort for AWS SA Bootcamp is launching on Oct 17th, 12 PM ET (Eastern Time) via live workshop. This program now includes our updated GenAI curriculum. Please register below:

Here’s what you get when you show up LIVE:

1. The myths keeping most people stuck - and what actually gets you hired as an SA - I've conducted over 300 SA interviews, so I know what I'm talking about!
2. How GenAI is reshaping the SA and Gen AI roles including FDE, and the exact AI concepts (RAG, agents, MCP, eval etc.) you need to speak fluently in interviews.
3. A first look at my new product feature, built to help you practice real-world, interview-relevant hands-on work instead of copy-paste tutorials.
4. Full bootcamp breakdown for Cohort 10, plus a special offer only for live attendees.
5. My exclusive Solutions Architect framework to prep you for today's job market! But if you’re not live, you won’t get it. No second chances.

And good news - it already worked for last cohort's students who secured cloud jobs in top companies, including at AWS, Microsoft, Google, JPMorgan, Reddit, and some of them didn't even have cloud experience 💰.

Spots are limited, so don't miss it!