Forward-deployed engineer
Works with a customer under real constraints to diagnose a workflow, deliver a credible technical wedge, and leave behind an operable system.
Also observed as · FDE · AI deployment engineer · customer-facing AI engineer
The trial is a synthetic operating decision, not a capability score. This advanced training route is not an employer screen, placement promise, or claim about your current readiness.
- Target standing
- L3
- Dependency gate
- 22 modules
- Authored module time
- Role proof
- The enterprise AI control plane
Module time excludes project build, review, and revision time.
From an ambiguous customer problem to an operable handoff.
Two layers work together. The 22-module role-entry route below composes the existing standard across five tracks. The additive eight-module FDE specialization is the seventh track: it deepens field execution without silently rewriting the frozen role gate.
- Field stages
- 6
- Route modules
- 22
- Authored time
- 15 hr 32 min
- Field simulations
- 3
- 01FIELD STAGE
Discover
Which customer workflow is valuable, bounded, and ready for evidence?
A defensible opportunity frame grounded in model limits, the available stack, and explicit operating economics.
ROUTE EVIDENCEThinking in models · provisioned lesson →The AI native tool stack · provisioned lesson →Prompting as specification · provisioned lesson →Designing your first AI workflow · provisioned lesson →Your personal AI operating system · provisioned lesson →The AI native landscape · provisioned lesson →Where AI actually pays · provisioned lesson → - 02FIELD STAGE
Frame
What is the smallest safe delivery wedge, and what must it prove?
A staged specification with an evidence bar, risk tier, owner, and decision boundary.
ROUTE EVIDENCE - 03FIELD STAGE
Build
How will the team turn the specification into an inspectable system?
A tested agentic implementation whose tool contracts, orchestration, and machine-written changes can be reviewed.
- 04FIELD STAGE
Deploy
What evidence must hold before this system reaches real work?
A bounded release with exact-candidate evaluation, canary evidence, rollback, and an explicit operating gate.
- 05FIELD STAGE
Recover
How will the team contain failure, preserve truth, and earn the right to resume?
An incident response that invalidates stale evidence, repairs the controlling defect, reruns the gates, and records the limitation.
- 06FIELD STAGE
Handoff
Who owns the system after the forward-deployed engineer leaves?
An operable control plane, named ownership, and a handoff contract the customer can continue without the embedded engineer.
One proof bar. Three honest starting positions.
The assessment and reviewed evidence can change where the work begins. They never waive the dependency or proof requirements.
Build from first principles
New to AI-native engineering or still assembling the technical foundation.
Route: Complete the dependency-closed 22-module route from L0 through the L3 role proof.
This is the full route. Completion and reviewed evidence—not elapsed time—determine progress.
Thinking in models →Diagnose the gaps
Already shipping software or AI systems and able to bring inspectable work evidence.
Route: Begin with the capability assessment, then use reviewed evidence to prioritize the dependency-closed route.
Assessment can adapt sequence and emphasis. It does not promise a skip, lower a gate, or waive the role proof.
Capability assessment and evidence review →Practice under field pressure
Experienced operator ready to rehearse recovery, release, governance, and customer handoff decisions.
Route: Use the advanced simulations as practice alongside any route gaps identified by assessment and evidence review.
The deterministic capstone is synthetic practice. Its replay receipt does not create graph progress and cannot replace the reviewed sdlc/control-plane role proof.
BME-10 · Run the forward-deployed field loop →Run the complete field loop before customer access.
BME-10 is the full deterministic capstone; BME-05 and BME-06 isolate recovery and governance decisions. All three use fictional packets and published standards. They create practice evidence—not customer experience, a hiring signal, or the role proof.
Run the forward-deployed field loop
Complete one inspectable Discover → Frame → Build → Deploy → Recover → Handoff loop and produce a content-addressed private evidence receipt without claiming real customer work, hiring readiness, curriculum completion, or credential evidence.
Recover the enterprise copilot pilot
Diagnose the failure mode, redesign the full-stack path and controls, define measurable release gates, and sequence a two-week recovery that earns adoption without claiming the pilot is ready today.
Gate a networked shell agent
Contain the run, determine what the evidence does and does not establish, and design the smallest governed path to an authorized production release.
A role target with a visible entry and a fixed bar.
The published route begins at L0 with Thinking in models and carries every prerequisite through the L3 target. The assessment can change where you begin; it does not change the proof required.
No module credit is assumed in this public route. Bring prior evidence to the assessment; qualifying reviewed work is the basis for any individual capability claim.
AI literacy & judgment
Translates model limits, risk, and economic tradeoffs into a decision a customer can make.
Prompting & instruction design
Uses precise instructions and tool contracts to make a customer workflow reliable enough to demonstrate.
Workflow decomposition
Finds the highest-leverage constrained workflow and turns it into a staged delivery plan with a measurable first win.
Agent direction & oversight
Selects an agentic pattern only when it improves the workflow and preserves human control.
Verification & quality control
Defends the delivered system with task tests, observability, and an operational response to failure.
Security & governance awareness
Works inside customer data, authorization, and risk constraints without treating compliance as someone else's handoff.
Every prerequisite. In learning order.
22 authored modules across 5 tracks. The 15 hr 32 min total is the curriculum estimate, not a promise of elapsed time to capability.
- FND2 hr 59 min
- AI Native Foundations
- 5 modules
- DEV3 hr 42 min
- AI Native Developers
- 5 modules
- AGI2 hr 46 min
- Agentic Workflows & AI Agents
- 4 modules
- EXE1 hr 51 min
- Executives & Leaders
- 3 modules
- SDL4 hr 14 min
- AI Native SDLC & Control Plane
- 5 modules
- 01 · L0Thinking in modelsFND · Calibration log: 3 prompt variants, drift analysis, and your trust rules.27 MIN
- 02 · L0The AI native tool stackFND · One-screen map of your AI stack with job labels and two identified gaps.33 MIN
- 03 · L0Prompting as specificationFND · Before/after prompt pair with constraint analysis on three test inputs.38 MIN
- 04 · L1Designing your first AI workflowFND · Workflow definition + two run logs + checkpoint analysis.43 MIN
- 05 · L1Your personal AI operating systemFND · One-page personal AI OS + three days of retro notes.38 MIN
- 06 · L2The AI native dev environmentDEV · AGENTS.md + agent-produced diff + failure analysis.37 MIN
- 07 · L2Spec-driven development with agentsDEV · Executable spec + agent plan + green diff.47 MIN
- 08 · L2Anatomy of an agentAGI · Minimal agent config + annotated execution trace.31 MIN
- 09 · L2Tools, MCP, and protocolsAGI · Tool contract v1→v2 with documented agent failure.36 MIN
- 10 · L2The AI native landscapeEXE · Scored vendor pitch + two forcing questions.31 MIN
- 11 · L2Where AI actually paysEXE · Scored use-case portfolio + top-three defense.37 MIN
- 12 · L2The AI native SDLCSDL · Eight-phase delivery map for a real agent workstream + gate definitions + accountability labels.46 MIN
- 13 · L2Risk tiers and the control matrixSDL · Risk-tier control matrix for five real systems + the design change classification forced.41 MIN
- 14 · L3Test-driven agentic codingDEV · TDD transcript with mutation-test evidence.41 MIN
- 15 · L3Reviewing machine codeDEV · Risk-tiered review of an agent PR with demanded changes.36 MIN
- 16 · L3Orchestrating multi-step agentsAGI · Three-agent orchestration + run evidence from both gates.51 MIN
- 17 · L3Evaluating agent behaviorAGI · Three-layer eval suite + adversarial fix.48 MIN
- 18 · L3Ship an agent to productionDEV · Working agent + consequence/coverage/variance release evidence + redacted event contract + measured cost sheet + exercised containment and rollback runbook.61 MIN
- 19 · L3The verifier layerSDL · Six-class verifier suite wired into a pipeline + signed results for one real change + an induced failure.55 MIN
- 20 · L3Agent observabilitySDL · Instrumented agent + one full trace + three runtime detection rules + a forensic reconstruction.48 MIN
- 21 · L3The enterprise AI control planeSDL · Control-plane design + a registered agent with scoped delegated authority + a policy denial + the orphaned-owner demonstration.ROLE PROOF64 MIN
- 22 · L3Directing agentic deliveryEXE · Delegation contract for one agentic workflow, with rung defense.43 MIN
The enterprise AI control plane
Control-plane design + a registered agent with scoped delegated authority + a policy denial + the orphaned-owner demonstration.
- NODE
- n-l3-11
- STANDING
- L3
- ALIGNMENT
- AINO-G · Certified Governed AI Delivery Engineer
Completion alone is not verification. Only qualifying reviewed work can support the evidence record; credential requirements remain mechanically derived from the Capability Graph.
A customer-style discovery and delivery brief
It separates the problem, stakeholders, constraints, success metric, staged scope, and decisions that need customer ownership.
Produced through
A deployed workflow with a handoff plan
The system is observable, has an operator path, and identifies what remains with the customer after the engineer leaves.
Produced through
A risk and verification record
A reviewer can see authorization, test evidence, escalations, and the operating owner for the delivered slice.
Produced through
Published availability, without a sales fiction.
These are the current Repo-backed delivery formats aligned to the role proof track. A schedule or staffing commitment exists only when its published record says so.
Governed AI Delivery — Cohort 01
- Schedule
- Schedule follows the founding cohort
- Cadence
- Weekly live · 10 weeks
The route names what it does not create.
Live customer discovery, procurement navigation, and executive-room facilitation require supervised field practice beyond the current modules.
These are authored role targets already carried by the LockedIn Labs curriculum Repo—not job listings, a labor forecast, an employer screen, or a placement promise. They are not a measurement of any learner. Dimension floors describe the target; only qualifying reviewed work can support an individual capability claim.
The scan does not claim every forward-deployed role carries the same travel expectation, clearance requirement, or industry specialization.
Test the operating standard before choosing the full route.
The public Operator Trial compresses one synthetic release decision into about ten minutes. It creates no learner score, record, or hiring signal.