NO COHORT HAS RUN
The runway is an authored operating model. It contains no completion, placement, qualification, or outcome rate.
[ FDE FORMATION · OPEN BLUEPRINT ]AUTHORED · NOT YET OPERATED
SELECTED TO TRAIN · PROVEN TO DEPLOY
See the whole journey from provider fluency to field entrustment.
Selected public Claude Academy courses and other external provider learning establish the product foundation under issuer ownership. The formation system then adapts prerequisite instruction, moves a cohort through business discovery and production-shaped builds, requires individual evidence inside team missions, adds supervised field work, and ends in an independent defense and a separate governed decision.
Independent curriculum crosswalk · course materials, badges, and certifications remain under Anthropic · no partnership, license, authorization, or endorsement implied.
ONE ROUTE · FIXED BAR
- 12–24 weeks
- Formation window
- 11
- Selected provider foundations
- 0–22
- Adaptive bridge
- 8
- FDE core modules
- 3
- Simulation boards
- 10
- Conjunctive gates
[ Formation mission deck ]Sequence · not learner progress
One operating sequence. Different kinds of proof.
The deck keeps the complete formation in view while each chapter moves from provider knowledge into enterprise decisions, production-shaped work, supervised evidence, and independent review.
Deck position
0108
00Open blueprint
You are viewing · Blueprint
See the complete formation system
What must be learned, built, observed, reviewed, and defended before deployment?
- Produces
- Formation runway and evidence map
- Authority
- Open authored blueprint
Deck position shows where this public chapter sits in the authored system. It is not saved learner progress, proof of completion, a cohort record, or a qualification decision.
[ FDE-RWY/0.1 ]AUTHORED · NOT ACTIVE
12–24 WEEK FORMATION MAP
See every move from foundation to field community.
The cohort advances on evidence, not elapsed time. Provider fluency opens the route; individual work, pod delivery, mentor observation, live boards, supervised field evidence, and an independent defense make the bar visible.
RUNWAY INSTRUMENT
- FOUNDATION
- 11 provider courses
- BRIDGE
- 0–22 modules
- FDE CORE
- 8 modules
- BOARDS
- 3 simulations
- BAR
- 10 evidence gates
STAFFING REQUIRED
Live boards, mentor observation, and supervised client work activate only with named, authorized coverage and a real mission boundary.
NOT OPERATIONAL
No configured live room, provider, token delivery, presence, consent, recording, or retention workflow is represented here.
PLANNED · NON-ISSUING
The complete route can make evidence reviewable; it cannot issue FDE-P while governed qualification authority remains inactive.
READ THE LANES
One cohort. Four visibly different kinds of work.
Team delivery matters, but no pod can hide an individual gap. Human observation matters, but presence is not proof. The runway keeps each learning mode distinct from the evidence and decision that permit advancement.
Individual work
Named contribution and judgment remain inspectable inside every team mission.
Pod work
Small teams integrate across roles without allowing group output to hide individual gaps.
Mentor moment
A qualified human observes a consequential decision, challenges it, and records the bounded evidence.
Live board
A changing case is defended in real time; participation alone never satisfies a gate.
THE FIXED BAR
Ten gates. None averaged away.
The gate rail shows where each standard enters the formation. Instruction can adapt; the terminal evidence contract cannot.
- EG-00
Unassisted baseline work sample
diagnostic
FIND ON RUNWAY - EG-01
Discovery and outcome contract
performance
FIND ON RUNWAY - EG-02
Enterprise integration slice
performance
FIND ON RUNWAY - EG-03
Provider-portable system
performance
FIND ON RUNWAY - EG-04
Evaluation and reliability control
performance
FIND ON RUNWAY - EG-05
Security and authority review
live-defense
FIND ON RUNWAY - EG-06
Production, adoption, and incident command
performance
FIND ON RUNWAY - EG-07
Supervised field mission
supervised-field
FIND ON RUNWAY - EG-08
Ownership transfer and field learning
live-defense
FIND ON RUNWAY - EG-09
Integrated qualification defense
live-defense
FIND ON RUNWAY
- 01BEFORE + WEEKS 01–02AUTHORED ROUTEExternal foundations
Provider foundation + unassisted baseline
Build product vocabulary through issuer-owned learning, preserve the external record, then establish an unassisted baseline before instruction changes the signal.
GATES ON THIS ROWEG-00FORMATION LANESIndividual work
Complete the selected external foundation and produce the original baseline response under recorded conditions.
Mentor moment
Separate exam preparation from field capability and identify the first observable gap without awarding standing.
External foundations- EXTERNALOPEN
AI Fluency for Builders
Anthropic remains the course or credential owner
- EXTERNALOPEN
Claude Platform 101
Anthropic remains the course or credential owner
- EXTERNALOPEN
Building with the Claude API
Anthropic remains the course or credential owner
- EXTERNALOPEN
Claude Code 101
Anthropic remains the course or credential owner
- EXTERNALOPEN
Claude Code in Action
Anthropic remains the course or credential owner
- EXTERNALOPEN
Introduction to Model Context Protocol
Anthropic remains the course or credential owner
- EXTERNALOPEN
Model Context Protocol: Advanced Topics
Anthropic remains the course or credential owner
- EXTERNALOPEN
Introduction to subagents
Anthropic remains the course or credential owner
- EXTERNALOPEN
Introduction to agent skills
Anthropic remains the course or credential owner
- EXTERNALOPEN
Claude with Amazon Bedrock
Anthropic remains the course or credential owner
- EXTERNALOPEN
Claude with Google Cloud's Vertex AI
Anthropic remains the course or credential owner
EVIDENCE TO MOVE- Issuer-owned learning or credential record
- Unassisted work sample with task, time, tool, and assistance conditions
- Gap map tied to published capability domains
ADVANCE WHENThe baseline is complete enough to locate strengths and gaps. Provider completion never substitutes for the remaining gates.
EXTENDED RUNWAYAdd targeted provider practice or repeat the diagnostic under valid conditions; do not manufacture a pass threshold for a diagnostic gate.
- 02WEEKS 01–04 · ADAPTIVEAUTHORED ROUTEAdaptive bridge · 0–22 modules
Close only the prerequisite gaps the evidence reveals
Use assessment and reviewed prior work to sequence a dependency-closed bridge of zero to 22 modules. Adapt the route, never the bar.
This row develops or maintains capability. It does not create a separate evidence gate.
FORMATION LANESIndividual work
Complete assigned modules and show the named skill in an inspectable artifact.
Pod work
Run short clinics where stronger members coach reasoning without completing another learner's proof.
Mentor moment
Review prior evidence, preserve prerequisites, and update emphasis without waiving a later gate.
Adaptive bridge · 0–22 modulesInspect the dependency-closed bridge
- 01.01OPEN
Thinking in models
Discover · AI Native Foundations
- 01.02OPEN
The AI native tool stack
Discover · AI Native Foundations
- 01.03OPEN
Prompting as specification
Discover · AI Native Foundations
- 01.04OPEN
Designing your first AI workflow
Discover · AI Native Foundations
- 01.05OPEN
Your personal AI operating system
Discover · AI Native Foundations
- 01.06OPEN
The AI native landscape
Discover · Executives & Leaders
- 01.07OPEN
Where AI actually pays
Discover · Executives & Leaders
- 02.01OPEN
The AI native dev environment
Frame · AI Native Developers
- 02.02OPEN
Spec-driven development with agents
Frame · AI Native Developers
- 02.03OPEN
Anatomy of an agent
Frame · Agentic Workflows & AI Agents
- 02.04OPEN
Tools, MCP, and protocols
Frame · Agentic Workflows & AI Agents
- 02.05OPEN
The AI native SDLC
Frame · AI Native SDLC & Control Plane
- 02.06OPEN
Risk tiers and the control matrix
Frame · AI Native SDLC & Control Plane
- 03.01OPEN
Test-driven agentic coding
Build · AI Native Developers
- 03.02OPEN
Reviewing machine code
Build · AI Native Developers
- 03.03OPEN
Orchestrating multi-step agents
Build · Agentic Workflows & AI Agents
- 04.01OPEN
Evaluating agent behavior
Deploy · Agentic Workflows & AI Agents
- 04.02OPEN
Ship an agent to production
Deploy · AI Native Developers
- 05.01OPEN
The verifier layer
Recover · AI Native SDLC & Control Plane
- 05.02OPEN
Agent observability
Recover · AI Native SDLC & Control Plane
- 06.01OPEN
The enterprise AI control plane
Handoff · AI Native SDLC & Control Plane
- 06.02OPEN
Directing agentic delivery
Handoff · Executives & Leaders
EVIDENCE TO MOVE- Assessment and reviewed-evidence gap map
- Dependency-closed assigned route
- Individual artifacts for every assigned gap
ADVANCE WHENEvery assigned prerequisite gap has inspectable evidence; acceleration changes instructional volume, not FDE evidence requirements.
EXTENDED RUNWAYKeep the learner on targeted clinics, revisions, and changed-case checks until the missing prerequisite is observable. The cohort need not move in lockstep here.
- 03WEEKS 02–10AUTHORED ROUTEFDE core · 8 modules
Traverse the eight-module FDE core
Move from business diagnosis through architecture, integration, portability, evaluation, secure authority, production command, and ownership transfer.
GATES ON THIS ROWEG-01EG-02EG-03FORMATION LANESIndividual work
Own the decisions and artifact slices attributed to you; every critical criterion remains individually inspectable.
Pod work
Integrate discovery, product, engineering, security, evaluation, and adoption work into one mission chain.
Mentor moment
Change a fact, challenge the authority envelope, and require revision before dependent work opens.
FDE core · 8 modules- FDE-01OPEN
Diagnose the work, establish the outcome
FDE core · N-L2-11
- FDE-02OPEN
Scope and architect the mission
FDE core · N-L2-12
- FDE-03OPEN
Land inside enterprise systems
FDE core · N-L3-13
- FDE-04OPEN
Engineer provider-portable AI systems
FDE core · N-L3-14
- FDE-05OPEN
Operate an evaluation and reliability control system
FDE core · N-L3-15
- FDE-06OPEN
Bound authority and earn authorization
FDE core · N-L3-16
- FDE-07OPEN
Operate adoption, rollout, and incidents
FDE core · N-L4-6
- FDE-08OPEN
Yield ownership and compound field learning
FDE core · N-L4-7
EVIDENCE TO MOVE- Sourced opportunity and outcome contract
- Executable enterprise integration wedge
- Unchanged behavioral contract across two provider or runtime configurations
ADVANCE WHENThe individual evidence for EG-01 through EG-03 meets every published pass condition; strengths cannot average away a critical miss.
EXTENDED RUNWAYReturn to the exact failed criterion, revise the artifact, and face a changed case. Additional time is support, not a lowered bar.
- 04WEEKS 06–12STAFFING REQUIREDSimulation boards · 3 practice environments
Rehearse consequence across three simulation boards
Run deterministic technical environments, then add live human pressure: regression, adversarial authority, changing operations, executive objections, and a defensible go, hold, repair, or stop decision.
GATES ON THIS ROWEG-04EG-05EG-06FORMATION LANESIndividual work
Operate the environment and preserve your own trace, decision, remediation, and rerun evidence.
Pod work
Rotate operator, technical lead, security, customer, and incident-command responsibilities.
Mentor moment
Inject a changing constraint and observe whether the engineer narrows, escalates, recovers, or stops appropriately.
Live board
Defend evaluation, authority, operations, and adoption decisions without scripts before active human coverage can accept a gate.
Simulation boards · 3 practice environments- BME-05OPEN
Recover the enterprise copilot pilot
Practice-only · Authored under: developers/ship-an-agent
- BME-06OPEN
Gate a networked shell agent
Practice-only · Authored under: developers/ship-an-agent
- BME-10OPEN
Run the forward-deployed field loop
Practice-only · Authored under: developers/ship-an-agent
BME-10 stays on the Formation Runway as a practice board authored under its existing module. It creates no Capability Graph progress.
EVIDENCE TO MOVE- Versioned evaluation and regression decision trail
- Threat, identity, data, and authority defense
- Staged release, incident containment, recovery, rerun, and adoption record
ADVANCE WHENEG-04 through EG-06 clear through their required evidence and active human review. A deterministic receipt alone is practice evidence, not gate acceptance.
EXTENDED RUNWAYReplay the technical environment, rehearse the live pressure, repair the controlling weakness, and rerun every invalidated gate under a changed condition.
- 05WEEKS 13–24 · WHEN ACTIVATEDSTAFFING REQUIREDField sequence
Earn supervised field evidence inside a real boundary
Own one bounded mission only when a named client decision owner, qualified supervisor, escalation path, evidence rights, and delivery agreement exist.
GATES ON THIS ROWEG-07EG-08FORMATION LANESIndividual work
Lead named decisions and preserve an attributable contribution map across the complete mission arc.
Pod work
Deliver as a small field unit while keeping client evidence isolated, permission-bound, and role-attributed.
Mentor moment
Retain escalation authority and record observed judgment, communication, recovery, and technical ownership—not attendance.
Live board
Review milestone evidence with the client owner and supervisor; media or presence never becomes proof by itself.
Field sequence- EG-07
Supervised field mission
Named supervisor + client decision owner required
- EG-08
Ownership transfer and field learning
Owner-executed change and recovery required
EVIDENCE TO MOVE- Client-ratified mission and milestone decisions
- Exact candidate-owned production release plus real-user or real-operator outcome, incident, guardrail, cost, and adoption evidence
- Verified rollback or recovery plus named-owner operation, change, teach-back, and accepted transfer
ADVANCE WHENBoth supervised-field and ownership-transfer gates are accepted from current, permission-bounded evidence by the required human authorities, including a separate real production release. Simulation or production-equivalent rehearsal can prepare the work but cannot satisfy the production-experience gate.
EXTENDED RUNWAYContinue bounded supervision, add another mission slice, or hold at the current authority level until the missing behavior is observed. Calendar time alone never creates entrustment.
- 06TERMINAL REVIEW · NOT ACTIVENOT ACTIVETerminal review
Defend the complete evidence chain independently
Explain business value, architecture, provider choice, evaluation, authority, operations, failure, adoption, and transfer while an independent panel changes one consequential condition.
GATES ON THIS ROWEG-09FORMATION LANESIndividual work
Defend the indexed record personally; a team artifact cannot answer for an individual's judgment.
Mentor moment
A prior mentor may clarify observation provenance but cannot replace independent panel judgment.
Live board
The panel tests adaptation across technical, risk, operator, and executive resolutions under a fixed, non-compensatory standard.
Terminal review- EG-09
Integrated qualification defense
Independent qualified human panel required
EVIDENCE TO MOVE- Indexed evidence chain across all capability domains
- Live scope change and injected failure response
- Panel decision with conflicts, holds, remediation, and appeal path recorded
ADVANCE WHENEvery conjunctive gate is current and accepted by an active governed authority. That authority is not active, so the runway cannot currently issue FDE-P.
EXTENDED RUNWAYA hold names the failed gate and evidence needed next. Return to supervised work or a changed-case defense; never force-rank the cohort or average over unsafe failure.
- 07ONGOING · PERMISSIONED MODELPERMISSIONED MODELGuild practice
Keep the cohort together as a Field Guild
Continue through permissioned case clinics, architecture boards, incident reviews, provider comparisons, and standard refreshes without turning community participation into standing.
This row develops or maintains capability. It does not create a separate evidence gate.
FORMATION LANESPod work
Original cohorts remain a durable practice network and form cross-cohort teams around permissioned problems.
Mentor moment
Graduates coach or review only after the applicable authorization, calibration, and conflict check.
Live board
Current-practice reviews can refresh evidence but cannot silently extend or issue a credential.
Guild practice- CLINIC
Permissioned case clinic
Client evidence stays inside its rights boundary
- REFRESH
Current-practice and standard refresh
Version changes can require new evidence
- PATTERN
Reusable field pattern
Provenance, permission, and known limits required
EVIDENCE TO MOVE- Permissioned case note with provenance and limits
- Peer critique and response record
- Current-practice refresh when the standard changes
ADVANCE WHENThere is no forced graduation from the community. Any standing remains governed separately by its current evidence and lifecycle rules.
EXTENDED RUNWAYIf evidence becomes stale or the standard changes, limit the claim and complete the named refresh before representing current standing.
[ FORMATION FIELD MANUAL · V1.0.0 ]PUBLISHED TEMPLATE · NO SAVED STATE
OPERATE THE POD · REVIEW THE WORK
Put the working agreement and the learning loop on the table.
A production-published, copy-and-complete operating agreement and after-action review protocol for bounded formation work. The kit is usable authored curriculum, not a saved cohort record or a runtime service.
- Release state
- Published template
- Reviewed
- Agreement
- OA-01
- Review loop
- 20 minutes
Markdown template · copy into an approved team workspace · no browser form or platform record is created.
01 · OA-01
Five owners. Rotating responsibility. No borrowed authority.
Complete and acknowledge at Muster before mission work begins. Review it whenever membership, roles, mission scope, authority, cadence, or escalation changes.
Completing or acknowledging OA-01 does not verify identity, attendance, consent, role performance, or competence; create an electronic signature, employment agreement, confidentiality agreement, client authorization, or legal obligation; or activate a cohort, mentor, decision log, credential, or runtime record. Applicable community rules, enterprise agreements, law, and named human authority prevail.
- R101 / 5
Mission Lead
Keeps the mission outcome, user, decision sequence, owner readbacks, dependencies, and escalation path coherent across the pod.
Authority limitCoordinates the mission but cannot unilaterally authorize production, accept organizational risk, score an evidence gate, or make a qualification decision.
- R202 / 5
Build Lead
Coordinates the production-shaped slice, implementation boundaries, typed interfaces, integration work, and reproducible build evidence.
Authority limitOwns build coordination, not every edit, and cannot widen tool, data, environment, or release authority.
- R303 / 5
Evals & Reliability
Coordinates evaluation specifications, held-out and adversarial cases, reliability signals, regression evidence, and rerun conditions.
Authority limitProduces and challenges evidence but cannot turn a score, model judgment, or green check into human release authority.
- R404 / 5
Security & Governance
Coordinates data, identity, authority, threat, control, residual-risk, retention, and escalation evidence for the mission.
Authority limitSurfaces and escalates governance decisions but does not become the client, legal, security, privacy, or risk-acceptance authority.
- R505 / 5
Operations & Adoption
Coordinates staged rollout, incident readiness, operator support, meaningful-adoption evidence, recovery, and owner transfer.
Authority limitCoordinates operating work but cannot infer adoption, attendance, delivery, or owner readiness from exposure or template completion.
OPERATING LOCKS
- L01Roles are rotating ownership assignments, not identities, status, attendance, or proof of competence. Every member holds Mission Lead at least once in the approved program sequence, while accepted observed evidence remains separately required.
- L02Every consequential decision is recorded in the working artifact with one named human owner, rationale, evidence reference, and review trigger. This template does not create a durable decision-log service.
- L03AI assistance is disclosed per artifact with the provider, tool, material contribution, and human verification stated without exposing prompts, credentials, secrets, or restricted context.
- L04Failures, near misses, blockers, and unresolved unknowns are reported with evidence and an owner; they are never hidden to protect a schedule, demo, score, or narrative.
- L05The pod adapts Recurse Center's four lightweight social rules with attribution: no well-actuallys, no feigned surprise, no backseat driving, and no subtle -isms.
- L06The pod references Recurse Center's three self-directives: work at the edge of your abilities, build your volitional muscles, and learn generously.
- L07Task disagreement is welcome and stays focused on the work, evidence, and decision. Lightweight social-rule repair does not replace the applicable LockedIn community rules, reporting path, or enterprise agreement for harassment, discrimination, safety, or misconduct.
- L08Shared fields contain no secrets, credentials, protected health information, direct personal identifiers, client names, unnecessary personal data, restricted source, or material that lacks permission to leave its original boundary. Approved private operational handles are allowed only where bounded role or action ownership requires them.
- L09Each significant stage or incident closes with AAR-01 before the next dependent gate; the review remains developmental and is never scored as competence.
02 · AAR-01
Reconstruct the event. Change the next attempt.
Run as soon as practical after the specific event and before the next dependent gate, while preserving safety, privacy, evidence permissions, and incident-response priority. Align the participants, discussion, and next action to the unit being trained: the pod reviews pod performance, while an individual review examines that individual's bounded performance without borrowing team results.
20MINUTES · SIX MOVES
[LOCKEDIN FORMATION AAR · V1]
- 0102 MIN
Frame
State the developmental, nonpunitive purpose; identify the pod or individual review unit, event boundary, intended standard, and ground rules.
- 0203 MIN
Intended
Reconstruct what was supposed to happen: outcome, task standard, authority boundary, stop condition, and accountable decision owner.
- 0305 MIN
Observed
Reconstruct what actually happened from objective media and participant perspectives; label unknowns rather than filling gaps from memory.
- 0404 MIN
Explain
Identify what to sustain, what differed, near misses, and causal conditions, focusing on the work, system, and roles rather than personal blame.
- 0504 MIN
Commit
Choose at most three changes, each with one owner, due date or review trigger, acceptance evidence, failure response, and reattempt condition.
- 0602 MIN
Read back
Confirm owned actions, unresolved unknowns, escalation, evidence permissions, and the next event that will test the change.
COMPLETION RULE
Complete means evidence-bearing—not merely discussed.
A completed formation AAR names one specific event and review unit, preserves developmental intent, actively includes the participant or participants, uses multiple information sources including at least one objective-media reference, and records at least one owned next action. If any element is absent, label the artifact "LIMITED REFLECTION — NOT A COMPLETED AAR"; never infer missing evidence.
WHEN THE RECORD IS INCOMPLETE
LIMITED REFLECTION — NOT A COMPLETED AAR
Missing participation, multiple sources, objective media, or an owned next action stays visible. Confidence and consensus never fill the gap.
RESEARCH & DOCTRINE
The scaffold is sourced. The outcome claim stays closed.
Reconstruct the event from multiple information sources, including at least one permissioned objective-media reference such as a trace, log, diff, ticket, test result, decision record, timeline, or recording. Memory alone cannot complete AAR-01.
- META-ANALYSISHuman FactorsDo Team and Individual Debriefs Enhance Performance? A Meta-Analysis
Defines a debrief through active self-learning, developmental intent, a specific event, and multiple information sources, and keeps general appraisal or solitary reflection outside that definition.
The reviewed studies span other training and work settings; this source does not establish LockedIn program effectiveness or transfer to FDE field performance.
- META-ANALYSISJournal of Applied PsychologyA meta-analysis of the effectiveness of the after-action review (or debrief) and factors that influence its effectiveness
Supports aligning the review to the individual or team, using objective performance-review media, and using a self-led approach for team-aligned reviews while treating facilitation and structure as interacting design choices.
The aggregate findings do not prove this protocol was run, improved a LockedIn cohort, or established any learner's competence.
- OFFICIAL DOCTRINEUnited States Department of the ArmyFM 7-0 Training, Appendix K: After Action Reviews
Provides the intended-versus-actual, sustain-or-improve, and next-performance sequence; open questions; participant self-discovery; standards-based discussion; objective observations; and recorded next actions.
Army training doctrine informs the scaffold but does not authorize LockedIn to represent military equivalence, endorsement, or FDE outcome validity.
[ ONE FIRM · ONE DELIVERY SYSTEM ]FORMATION → FIELD
The training system prepares the engineer. The firm owns the mission.
LockedIn Labs is an implementation, consulting, and delivery firm—not a software marketplace. A named human board selects the cohort; the training system forms practitioners. SprintLoop FDE is the proprietary field tool being developed to support work inside separately activated client engagements.
ACCOUNTABLE OPERATORFIRM-CENTERED
AI implementation, consulting, and delivery firm
AUTHORED CONTRACT · NOT AN OPERATED PROGRAM
Under an activated client engagement, LockedIn Labs is the accountable implementation counterparty. The training system develops and tests the practitioner; SprintLoop FDE is the proprietary field tool being developed to structure delivery; neither replaces client authority or accountable human judgment.
- Assessment criteria
- 15
- Capability dimensions
- 6
- FDE role floors
- 6
- Formation phases
- 6
- Evidence gates
- 10
- Mission workstreams
- 6
THE OPERATING SEQUENCE
Build capability. Entrust narrowly. Deliver responsibly.
This five-move bridge connects formation to delivery; it is not a replacement name for the DEPLOY competency arc or SprintLoop FDE's engagement lifecycle. Training evidence informs a human decision; it never silently turns into employment, qualification, staffing, or customer access.
- 01SELECT
Named LockedIn human selection board
Admit, redirect, hold, or decline against the role contract and recorded work—not course history alone.
Baseline work sample, six-dimension role floors, stated conditions, provenance, and an accountable human decision.
- 02FORM
Training faculty and program operations
Advance, remediate, or hold through increasingly consequential practice and mission-specific preparation.
Versioned curriculum, artifacts, revisions, reviews, simulations, mission pack, and provider foundations.
- 03ENTRUST
Authorized reviewers, supervisors, and decision authority
Permit only the next bounded responsibility supported by observed evidence; never infer independent readiness from completion.
Conjunctive gates, supervised mission records, recovery evidence, owner attestation, independent defense, and unresolved holds.
- 04DEPLOY
LockedIn Labs delivery leadership
Staff and govern the client mission under its own scope, access, security, outcome, supervision, and escalation contract.
Minimal readiness envelope plus a newly authorized engagement record inside the proprietary field operating system.
- 05IMPROVE
Method and capability governance
Promote only permission-bounded, reproducible field learning into a new method, mission-pack, or evaluation version.
Provenance, client rights, de-identification, review, version decision, and explicit rejection or supersession lineage.
SELECTION CONTRACT
Select from demonstrated work—not résumé vocabulary.
The role target sets a floor for every capability dimension, but no composite score makes the decision. Selection combines a bounded baseline, an evidence interview, a repair cycle, and a recorded human judgment with explicit conditions and redirects.
ROLE · Forward-deployed engineerFDE-S/1.0 · V1.0.1
- AI literacy & judgment
- Floor 3/4W2
- Prompting & instruction design
- Floor 3/4W2
- Workflow decomposition
- Floor 4/4W3
- Agent direction & oversight
- Floor 3/4W2
- Verification & quality control
- Floor 3/4W3
- Security & governance awareness
- Floor 3/4W3
- S1
Eligibility and operating context
Program eligibility, mission-relevant experience, availability, program conditions, and declared assistance conditions. Accommodation records remain access-separated and non-evaluative.
Administrative screen only. Accessibility needs are administered by authorized staff and are never disclosed to evaluative scorers or used as capability evidence.
- S2
Bounded baseline work sample
Three bounded working exercises completed without generative assistance under declared accessible conditions, then scored against the published fifteen-criterion, six-dimension instrument with provenance.
Diagnostic. A live model review informs the record but never makes the admission or hiring decision.
- S3
Role-fit and evidence interview
FDE role floors, artifact walkthrough, applied system design, evaluation, governance, and communication under ambiguity.
A named LockedIn selection board records the decision, reasons, conditions, conflicts, and redirect path after calibrated human review.
- S4
Formation trial and repair
A production-shaped task, feedback response, revision quality, failure recovery, and learning velocity under fixed conditions.
Supports cohort selection or remediation; it is not qualification, employment, or client authorization.
An independent reconsideration and appeal path may review evidence, process, conflicts, and accommodations without lowering the published capability bar.
Authorized program owners calibrate reviewers and monitor decision consistency and adverse impact before each selection wave advances.
Selection evidence has a declared purpose, access boundary, retention period, correction path, and deletion or legal-hold disposition.
The named LockedIn selection board—not the training system—makes the program decision. No single score, provider badge, demographic proxy, completion event, or generated recommendation admits, hires, qualifies, or deploys an engineer.
MISSION PACKS
Add industry depth without weakening the common FDE bar.
A mission pack is a versioned overlay—not a collection of extra courses. It binds domain sources, workflow evidence, controls, role-players, production-shaped rehearsal, and field activation to the same underlying capability standard.
- MP-01
Mission and outcome definition
Bind the workflow, baseline, target condition, guardrails, kill criteria, decision owner, and evidence window before solution design.
- MP-02
Industry source and control set
Version the authoritative domain sources, policy constraints, data classes, retention rules, and review date; label unresolved interpretation.
- MP-03
Workflow and stakeholder model
Map the real-shaped work, systems of record, exceptions, incentives, decision rights, operators, risk owners, and executive sponsor.
- MP-04
Production-shaped mission rehearsal
Provide a synthetic or authorized environment with injected failures, authority limits, rollback, adoption, and owner-transfer work.
- MP-05
Evidence and entrustment contract
Name the required artifacts, observers, review rubrics, critical holds, remediation path, and human authority for each decision.
- MP-06
Field-system activation boundary
Define what the proprietary field tool may receive, what must be reauthorized inside the client engagement, and what never crosses systems.
FORMATION → FIELD HANDOFF
Share the minimum decision envelope. Reauthorize everything else.
The handoff is a contract between two systems, not a database merge. The client engagement creates its own identity, access, scope, supervision, evidence, retention, and outcome authority.
SPECIFICATION ONLY · NO LIVE CONNECTOR
- Owns assessment conditions, curriculum version, artifacts, review provenance, remediation, supervised evidence, and entrustment recommendations.
- Produces a minimal, signed, revocable readiness envelope only after a separately authorized human handoff decision.
- Owns the client mission, outcome definition, interviews, workflow and process maps, opportunity decisions, delivery plan, client evidence, controls, and operating history.
- Reauthorizes identity, tenant, role, scope, data, tools, environment, supervision, and expiry inside every engagement; training standing grants no client access.
- Engagement-scoped, non-correlatable practitioner reference and exact role/standard version
- Purpose, authorized recipient, entrustment scope, conditions, expiry, unresolved holds, and named revocation authority
- Mission-pack code/version and permitted evidence-receipt digests
- Recorded notice and lawful basis or explicit consent, plus access, correction, retention, deletion, and legal-hold controls
- No raw learner artifact, accommodation record, or client content unless separately purpose-limited and authorized
- No shared login, silent account provisioning, or copied session token
- No automatic hiring, qualification, staffing, or production authorization
- No raw assessment response, learner artifact, reviewer identity, or customer evidence in the portable envelope
- No stable cross-client practitioner identifier, secondary use, recipient expansion, or indefinite retention
- No field outcome written back as training credit without a new consented and reviewed evidence record
BOUNDARY · lockedin.field-handoff/v1
This is a proposed method-and-evidence interface—not a merged product, database, identity plane, or claim that SprintLoop FDE is integrated or operating for a customer today. The DEPLOY practitioner competency arc and SprintLoop's engagement lifecycle remain distinct and require an explicit mapping contract.
ENTERPRISE DECISION STUDIOS
Find the mission. Then earn the decision.
Forward-deployed work begins before architecture. These two browser-based studios train the sequence from contradictory workflow evidence to an adoption-adjusted value case, accountable coalition, buying path, and executive decision memo. They are formative practice—not saved evidence, field work, or qualification.
- 01 · DIAGNOSEBROWSER PRACTICE
Enterprise Discovery Studio
Reconstruct a synthetic healthcare workflow, separate observations from assumptions, choose the smallest defensible wedge, and revise when a material fact changes.
Output: Workflow · bottleneck · outcome contract
Run the discovery studio - 02 · ESTABLISHBROWSER PRACTICE
Value & Coalition Studio
Underwrite value without calling capacity cash, compare the no-build path, map operational and buying authority, handle objections, and seal a revised executive memo.
Output: Economics · coalition · decision memo
Build the value case
PROVIDER RELATIONSHIP
External foundation. Independent standard.
- B01Claude Academy and other provider courses remain externally hosted and issuer-controlled; this crosswalk does not copy, license, or imply endorsement of their curriculum.
- B02Provider completion and certification records never substitute for LockedIn mission, field, defense, or transfer evidence.
- B03A cohort seat does not imply employment, client access, certification, qualification, or a lifetime service commitment.
- B04The alumni field guild is an authored continuity model; ongoing programming activates only with operating coverage and member consent.
NEXT INSPECTION
Inspect how supervised experience becomes reviewable evidence.
The Mission Logbook makes hours, responsibility, workstreams, observations, field evidence, remediation, and the qualification boundary inspectable without pretending that time alone creates readiness.