From Consultant or Solutions Engineer
You already read rooms well. The gap is the engineering bar, and it's not negotiable.
The sequence
- 01
Close the engineering gap honestly
This is the whole route. Customer skill without a real engineering bar produces a technical account manager, which is a good job and a different one.
- 02
Build one real thing end to end
Not a notebook, not a demo. Something with a data source you don't control, a service, an interface, tests, and a deployment. Then operate it for a month.
- 03
Get fluent in the current deployment stack
For 2026 that means cloud, one database you know well, containers, CI, and enough of the AI stack — retrieval, structured outputs, evaluation discipline, one agent framework — to be useful on the engagements being staffed right now.
- Framework axis: Technical Breadthframework
- 04
Convert relationship skill into technical authority
You can already run the room. The shift is being the person who also builds the thing, which changes how your recommendations are received.
- 05
Learn the loop back to product
Consulting output ends at the customer boundary. Forward deployment doesn't — what returns to the roadmap is where the model earns its cost.
The honest version of this route
You start ahead on the axes that engineers find hardest. Customer Context, Problem Framing, and Communication and Influence are the ones that wash strong engineers out of the role, and you've been doing all three professionally for years.
You also start behind on the one that's not negotiable.
Forward deployed engineers write production code. They commit to customer repositories, debug pipelines at 11pm, and are accountable for systems that run after they leave. A hiring loop for this role will test that directly, and no amount of customer skill compensates for failing it. Anyone telling you otherwise is selling a course.
This is the longer route. Twenty weeks is a realistic floor if you're starting from scripting-level ability, and only if you're building consistently rather than reading.
What actually closes the gap
Build one real thing and operate it. Not a tutorial project. Something with an uncooperative external data source, real authentication, a deployment, and a month of you being on the hook when it breaks. That single experience moves you further than any course, because it teaches the thing the role is actually about: what happens after the demo.
Don't skip fundamentals for framework fluency. The interview will find it. Data modelling, how systems fail, and how to debug something you didn't write are the parts that transfer to every engagement.
Use the advantage you've. In an interview loop, your ability to reframe a vague customer problem and drive a session to a decision is genuinely differentiating. Most candidates from the engineering side can't do it. Lead with it — after you've cleared the technical bar, not instead of.
Where you'll probably land
Most people making this transition enter at FDE I or FDE II on the customer-facing axes and FDE I on Technical Breadth, which averages to a real FDE I offer at a good company. That's not a demotion; it's a different ladder, and the customer axes carry you up it faster than they carry an engineer.
Take the assessment and see the shape of your own profile. The lopsidedness is the useful part.