If your AI initiative needs serious engineering help, you have three honest options: hire a full-time AI engineer, bring in a traditional consultancy, or embed a fractional forward deployed engineer. Each is right for a different situation, and I'll try to be straight about which — including when hiring beats engaging someone like me.
Three honest options when your AI initiative needs help: hire in-house (permanence, but months to productive in a thin market), a consultancy (a deliverable whose knowledge walks out), or a fractional forward deployed engineer (the experience gap filled now, with transfer built in). Which fits when — including when you should just hire.
Most stalled AI pilots aren't failing for the reason everyone in the room thinks. The model is usually fine. What's missing is almost always one of five unglamorous things — and knowing which one changes whether the pilot ships in six weeks or dies in six months of polite status meetings. Here's the diagnostic I run when a team brings me a pilot that won't move.
Most stalled AI pilots aren't failing for the reason everyone thinks — the model is usually fine. The five causes I actually find: no written bar for done, proof on data the real world doesn't send, under-scoped integration, an unretired stakeholder fear, and no owner. The diagnostic, and what unsticking takes.
The most valuable output of a forward-deployed engagement isn't the system you shipped — it's what you learned shipping it. Captured and fed back, those field lessons turn bespoke wins into reusable patterns and better products. This closing post maps the theme that makes the whole discipline compound.
The most valuable output of an engagement is what you learned shipping it. Captured and fed back, field lessons turn bespoke wins into reusable patterns and better products — the discipline that makes forward deployment compound instead of rotting into a custom-build shop. Lessons, reuse, patterns, product feedback, and the field-to-product flywheel; closes the six-theme field guide.
Everything technical in this field guide runs on something human: the ability to work with stakeholders, grow in the role, build a team, and shape a community. This theme is the craft and the career — the part that doesn't show up in a repo but decides whether the rest works. This post maps it.
Everything technical runs on something human: working with stakeholders, growing in the role, building a team, and shaping a community. The craft and the career — stakeholders, the role and spectrum, careers, hiring, mentoring and coaching, community and influence — plus why the role is exploding at AI companies and builder-communicators are the scarce resource.
Proving value on a prototype is the halfway point. The other half — and the harder one — is making it survive production: evaluated, secured, compliant, reliable, and owned by the customer's team. This post maps the shipping theme of the field guide, where most AI initiatives actually stall.
Proving value on a prototype is the halfway point; the harder half is making it survive production — evaluated, secured, compliant, reliable, and owned by the customer's team. Deployment, evaluation, AI security, sovereignty and compliance, reliability, and handover, plus the new AI topics (agent security, operationalized governance, sovereign AI, evals as infrastructure).
You can execute flawlessly on the wrong problem and create nothing. The highest-leverage work a forward-deployed engineer does is choosing what to build and proving, on real data, that it was worth it. This post maps the discovery-to-value theme of the field guide.
You can execute flawlessly on the wrong problem and create nothing. The highest-leverage work is choosing what to build and proving, on real data, that it was worth it: discovery, use-case selection, scope, prototypes and acceptance criteria, and value — plus the AI-era additions (eval-driven scoping, the POC as the crux, value measurement for probabilistic systems).
The AI era is quietly rewriting the technical-sales role. When a demo no longer proves value and the deal is decided in production, the person who wins it isn't the one who presents best — it's the one who can build the win next to the customer. This closing post is about that shift: from sales engineer to forward deployed engineer, the apex role of AI-era technical GTM.
When a demo no longer proves value and the deal is decided in production, the person who wins it is the one who can build the win next to the customer. The technical-GTM spectrum (SE → FDE → CS), why AI made the forward deployed engineer mandatory, the craft tension of scoping and productizing, and how the role becomes a career and a business.
This is the theme people picture when they hear "AI engineer" — the actual building of agents and AI systems. But a forward-deployed engineer builds differently: on the customer's real data, inside their constraints, with cost and latency treated as first-class from day one. This post maps the building theme of the field guide.
The technical heart of the discipline — building agents and AI systems — done the forward-deployed way: on the customer's real data, inside their constraints, with cost and latency first-class. Agents and orchestration, MCP and connectors, context engineering, data and grounding, cost and latency, and the new AI topics (MCP as a standard, agent memory, cost/latency engineering).
A forward-deployed engineer's first job isn't to build — it's to understand. Before a line of code, you have to see how the customer actually works, whether they're ready for AI, and what it will take for them to adopt it rather than admire it. This opens a field guide to forward-deployed engineering, organized into six themes, and starts where every engagement starts: with the customer.
A forward-deployed engineer's first job isn't to build — it's to understand the customer: how they really work, whether they're ready for AI, and what adoption will take. This opens a six-theme field guide to the FDE discipline and covers workflow mapping, AI readiness, adoption as a people problem, and the new AI dynamics (calibrated trust, workflow-native agents, AI maturity).
I'm opening up my consulting, focused on forward deployed engineering for AI and technical teams. Here's the work I'd love to help with — and why it matters to me.
I'm opening up my consulting, focused on forward deployed engineering for AI and technical teams. One of the most remarkable things about this moment in AI is how quickly an idea becomes a convincing demo — and the most meaningful work is turning that demo into dependable value inside a customer's real environment. Here's the work I'd love to help with, and why it matters to me.
The best outcome of a consulting engagement isn't that the consultant becomes indispensable — it's that the team no longer needs them. Here's the mindset that gets you there, and why it's the right way to do forward-deployed work.
The best outcome of a consulting engagement isn't that the consultant becomes indispensable — it's that the team no longer needs them. The multiplier mindset — amplifying a team's capability instead of creating dependency — is both the more honorable goal and the better business, and it's the heart of forward-deployed work done well.
Any single engagement can be won with cleverness. A career can only be built on something steadier: a reputation for leaving every team genuinely better off. This is the last essay in the series, and it's about the longest view I have.
Any single engagement can be won with cleverness; a career can only be built on a reputation for leaving every team genuinely better off. Reputation as the only durable asset, generosity as a long-term strategy, what stays when the technology changes, and the single standard that lets a career take care of itself.
Forward-deployed work is intense by nature: you embed, you care, you carry someone else's deadline as your own. Doing it once is exhilarating. Doing it for years without burning out takes a kind of discipline nobody warns you about. This is what I've learned about staying in the game.
Forward-deployed work is intense by nature — you embed, you care, you carry someone else’s deadline as your own — and doing it for years without burning out takes a discipline nobody warns you about. The specific intensity of embedded work, protecting the asset that is you, financial and psychological independence, and variety as renewal.
A consulting practice is not built by marketing. It's built by engagements that go so well the client becomes the source of the next one. Here's how a single good win quietly compounds into a practice that fills itself.
A consulting practice isn’t built by marketing — it’s built by engagements that go so well the client becomes the source of the next one. Why forward-deployed work refers so well, why the second engagement with the same client is the most valuable pipeline, how to tend the compounding without selling, and the patience it requires.
The handoff is the moment of truth for forward-deployed work. Everything you did or didn't do earlier shows up here, in whether the team can carry the thing forward without you. A good handoff isn't an event at the end; it's something you've been doing all along.
The handoff is the moment of truth for forward-deployed work — everything you did or didn’t do earlier shows up in whether the team can carry the thing forward without you. Why a good handoff starts on day one, handing over understanding rather than just artifacts, and how to engineer your own obsolescence on purpose.
You can deliver the same working system two ways: by building it yourself and handing it over, or by building it with the team so the capability stays. They look similar on the last day and could not be more different six months later.
You can deliver the same working system two ways — build it yourself and hand it over, or build it with the team so the capability stays. They look identical on the last day and could not be more different six months later. Why the black box is tempting and costly, what building alongside looks like daily, and how it redefines the deliverable.
The consultant who says yes to everything looks accommodating and is quietly failing the client. Learning to say no — to the wrong engagement, the wrong feature, the wrong architecture — is one of the clearest marks of someone worth trusting.
The consultant who says yes to everything looks accommodating and is quietly failing the client. Why the unexamined yes harms — to the wrong engagement, the wrong feature, the wrong architecture — how a well-reasoned no is what makes your yes trustworthy, how to say it well, and the hardest no of all: to work you want but shouldn’t take.
A forward-deployed consultant is a stranger given deep access to a team's work. Everything that follows depends on how quickly that stranger becomes someone the team is glad is there. The first two weeks decide it.
A forward-deployed consultant is a stranger given deep access to a team’s work, and the first two weeks decide whether that stranger becomes someone the team is glad is there. Building trust through small visible usefulness, pairing credibility with warmth, winning over the person who didn’t want you there, and why trust is the real deliverable of the beginning.
The temptation at the start of an engagement is to start building, because building is what you're good at. Almost every engagement I've seen go wrong went wrong here, by skipping the quiet work of understanding first. This is a defense of discovery.
The temptation at the start is to start building, because building is what you’re good at — and almost every engagement that goes wrong went wrong by skipping discovery. What discovery is actually for (how the work really happens, what the data is really like, what ‘good enough to trust’ means), how it earns you standing, and how to keep it from becoming a stall.
Engineers price consulting the way they bill time: by the hour. It's the most natural instinct and, I think, usually the wrong one. Here's how I've learned to think about pricing work whose value has nothing to do with how long it takes.
Engineers price consulting the way they bill time — by the hour — and it’s usually the wrong instinct. Why hourly billing ties your pay to your inefficiency and puts you on the wrong side of every improvement, how to price the outcome instead, why pricing is a trust signal, and some honest caveats about value-based pricing.
Almost every engagement starts as a wish, not a scope. The skill that most determines whether it succeeds is turning that vague ask into one bounded, achievable first win. Here's how I try to do it.
Almost every engagement starts as a wish, not a scope, and the skill that most determines success is turning that vague ask into one bounded, achievable first win. Scoping to a first win rather than the whole mountain, discovering before committing, writing the scope down in plain language, and leaving room for what you’ll learn.
The hardest client to find is the first one, and almost none of the advice about it is honest. Here's how forward-deployed consulting work really begins — not with cold outreach and a funnel, but with trust that already exists.
The hardest client to find is the first one, and most advice about it isn’t honest. Where forward-deployed work really begins — not with cold outreach and a funnel but with trust that already exists — what you’re actually selling, how to make yourself findable, and why you should start before you feel ready.
Most consulting either hands over a deck or rents out hours. Forward-deployed consulting does neither — it embeds a builder next to your team to ship a real result on your real data. Here's what that actually means, and why I've built my practice around it.
Most consulting either hands over a deck or rents out hours; forward-deployed consulting does neither — it embeds a builder next to your team to ship a real result on your real data. What the work actually is, how it differs from strategy and contracting, and why the demo-to-production gap in AI is exactly where it earns its keep.
A practical blueprint for founders and GTM leaders: when to start forward deployment, who to hire first, how to scope engagements, what to measure, and how to keep it from eating your margins.
A practical blueprint for founders: when to start forward deployment, who to hire first, how to scope engagements, what to measure, and how to keep it from eating your margins — plus the four failure modes (agency, hero, science-project, over-serving) and a 90-day start.
Forward deployment that isn't run as a disciplined function becomes an unscalable custom-dev shop. The Forward Deployed Leader owns the motion as a business — the team, the scoping, the economics, and the decision of what becomes product.
Forward deployment that isn't run as a disciplined function becomes an unscalable custom-dev shop. The Forward Deployed Leader owns the motion as a business — team design, scoping discipline, unit economics, and the decision of what becomes product — and names the invisible trap: it feels like success right up until the margins invert.
The forward deployed engineer builds a win inside one customer. The Forward Deployed Architect makes that win survive the next ten — turning bespoke builds into reference architecture, passing security review, and deciding what's reusable. It's the missing layer between heroics and product.
The forward deployed engineer builds a win inside one customer; the Forward Deployed Architect makes it survive the next ten — reference architecture, security and governance that passes review, and the reuse decision that separates product from one-off. The missing layer between heroics and product.
The forward deployed engineer gets all the attention, but forward deployment is bigger than one role. It's a go-to-market operating model with three layers — build, scale, run — and seeing the whole stack is how AI companies turn embedded engineering from a heroic cost center into a repeatable engine.
The forward deployed engineer gets all the attention, but forward deployment is bigger than one role — it's a go-to-market operating model with three layers: build (engineer), scale (architect), run (leader). Seeing the whole stack is how AI companies turn embedded engineering from a heroic cost center into a repeatable engine.
Every AI forward deployed engineer builds one-offs — a bespoke deployment for one customer's data, workflow, and trust. The ones who create lasting value turn those one-offs into product: the patterns that repeat become a platform, the platform makes the next deployment faster, and the field learnings flow back to shape what gets built. This closing post is about the flywheel that turns bespoke AI work into a compounding asset, and the career arc of the engineer who runs it.
Every AI FDE builds one-offs — a bespoke deployment for one customer's data, workflow, and trust. The ones who create lasting value turn those one-offs into product: the patterns that repeat become a platform, the platform makes the next deployment faster, and field learnings flow back to shape what gets built. The flywheel that turns bespoke AI work into a compounding asset — and the AI FDE career arc.
A pilot that works is not a system the customer can run. Productionizing AI means making it reliable, affordable, fast, and observable enough to be real infrastructure — and then handing it over so the customer operates it without you. The forward deployed engineer's goal, in the end, is to make themselves unnecessary: a deployment that only works while you're standing next to it hasn't actually been delivered.
A pilot that works is not a system the customer can run. Productionizing AI means making it reliable, affordable, fast, and observable enough to be real infrastructure — then handing it over so the customer operates it without you. The FDE's goal, in the end, is to make themselves unnecessary. Cost, latency, drift, observability, and a handover that transfers the eval discipline, not just the code.
A technically excellent AI system that nobody uses has delivered zero value. The last mile of an AI deployment is not the model — it's fitting the system into how real people actually do their jobs, designing an interface that handles uncertainty honestly, and managing the human change of introducing AI into someone's work. This is where deployments succeed or quietly fail, and where the forward deployed engineer's non-technical skills matter most.
A technically excellent AI system that nobody uses has delivered zero value. The last mile of an AI deployment is not the model — it's fitting the system into how real people actually do their jobs, designing an interface that handles uncertainty honestly, and managing the human change of introducing AI into someone's work. Human-in-the-loop, calibrated reliance, autonomy levels, and change management.
You cannot ship AI you cannot measure, and no enterprise grants a probabilistic system authority over real work on faith. Both problems have the same answer: evaluation. Building the customer's own evaluation set — real examples, their definition of correct — is how the AI forward deployed engineer turns "it seemed good in the demo" into a reliability number, and that number is how trust gets earned.
You cannot ship AI you cannot measure, and no enterprise grants a probabilistic system authority over real work on faith. Both problems have the same answer: evaluation. Building the customer's own eval set — real examples, their definition of correct — turns 'it seemed good in the demo' into a reliability number, and that number is how trust gets earned. Offline/online eval, safe failure, and trust.
A frontier model knows the public internet and nothing about the customer. All the value of an AI deployment comes from the opposite: making the model reason over the customer's own documents, records, and knowledge. Grounding is the technical heart of the AI forward deployed engineer's job — connecting a general model to a specific company's messy, permissioned, incomplete data so its answers are about their reality, not the model's imagination.
A frontier model knows the public internet and nothing about the customer. All the value of an AI deployment comes from the opposite: making the model reason over the customer's own documents, records, and knowledge. Grounding — retrieval-augmented generation over messy, permissioned, incomplete data — is the technical heart of the AI FDE's job. With an interactive reference-architecture diagram.
An AI demo is the easiest impressive thing to build and the most misleading. It runs on hand-picked inputs, in a clean environment, with the failures edited out — and it convinces everyone the problem is nearly solved when the real work has barely begun. The AI forward deployed engineer's job in this phase is to use the demo to win belief, then walk the customer honestly across the chasm to a pilot that survives real data.
An AI demo is the easiest impressive thing to build and the most misleading: it runs on cherry-picked inputs, in a clean environment, with the failures edited out — and convinces everyone the problem is nearly solved when the real work has barely begun. Use the demo to win belief, then walk the customer honestly across the chasm to a pilot that survives real data.
The most important decision an AI forward deployed engineer makes happens before any code: which problem to point the model at. Choose a problem AI is genuinely suited for, with real value and a clear way to measure it, and the engagement can succeed. Choose AI theater — impressive-sounding but ill-fit — and no amount of engineering saves it. Scoping is where AI deployments are won or lost.
The most important decision an AI FDE makes happens before any code: which problem to point the model at. Choose a problem AI is genuinely suited for, with real value and a clear way to measure it, and the engagement can succeed. Choose AI theater — impressive-sounding but ill-fit — and no engineering saves it. Fit vs value, resisting AI theater, augmentation over automation, and picking the wedge.
The forward deployed engineer was born at Palantir to bridge powerful software and messy customer reality. In the AI era the role has exploded, because frontier models have made that gap wider than ever: a model that dazzles in a demo is a long way from a system that works inside one company's data, workflows, and trust constraints. This series is the technical playbook for the engineer who closes that gap.
The forward deployed engineer was born at Palantir to bridge powerful software and messy customer reality. In the AI era the role exploded, because frontier models widened that gap: a model that dazzles in a demo is a long way from a system that works inside one company's data, workflows, and trust constraints. This series is the technical playbook for the engineer who closes that gap.
The role that puts you closest to real problems and real users also carries the sharpest burnout and career traps — thriving as an FDE means managing both deliberately.
The capstone: building a durable FDE career — the rewards and the hazards (context-switching, emotional labor, the maintenance trap), managing your capacity and boundaries, and growing toward staff engineering, product, leadership, or founding.
The forward deployed engineer's edge isn't deep mastery of one stack — it's enough breadth to build an end-to-end solution alone, fast, against whatever the customer already has.
The FDE's edge is breadth, not deep single-stack mastery: comb-shaped competence across data, backend, a little frontend, and just-enough ops; choosing tools for speed and fit; a pragmatic default kit; and the meta-skills (learning speed, finishing) that outlast any framework.
Every forward deployed engineer builds one-offs to win the customer in front of them — the ones who last turn those one-offs into product instead of drowning in them.
Every FDE builds one-offs; the ones who last turn them into product. Why bespoke is the right start, the 'third time productize' rule, the feedback loop to the product team, designing custom work for graduation, and managing the portfolio of one-offs.
An FDE's code only matters if the customer trusts them enough to adopt it — the relationship is not soft-skills garnish, it's the delivery mechanism.
An FDE's code only lands if the customer trusts them: earning trust in small kept promises, speaking the customer's outcomes and vocabulary, managing expectations relentlessly, navigating the org (champion/skeptic), and delivering bad news well.
The demo ran on your laptop with clean data — production means the customer's messy systems, their security rules, and their data as it actually is, which is where most forward deployed work is really won.
Production means the customer's messy data, systems you don't control, and security you must pass: profiling dirty data, loose-coupling integrations, treating security/residency as gates, and deploying + handing over so you can actually leave.
An FDE's superpower is turning a vague problem into something the customer can see and touch within days — because a rough working demo teaches more than a month of meetings.
Turn a vague problem into something the customer can touch in days: build the thinnest slice that tests the riskiest assumption, run a tight demo loop, and manage the prototype's lifespan so 'it demoed' doesn't get shipped as 'it's done'.
The problem a customer first describes is almost never the problem worth solving — an FDE's first job is to dig until the real one surfaces.
The problem a customer first states is rarely the one worth solving: discovery techniques (ask why, watch real work, find the decision), the jobs-to-be-done lens, mapping stakeholders and constraints, and writing a confirmed problem frame.
Part software engineer, part consultant, part product manager — the forward deployed engineer works inside the customer's world to turn a hard problem into working software, then carries what they learn back to the product.
The opener to a forward deployed engineering series: the FDE role as a blend of engineer, consultant, and product manager — embedded at the customer, building real software against a vague problem, then carrying the learnings back to the product.