The engineering pathway
A conversion programme for existing engineers. Volume 1 completion means foundations completed: a gated workflow in the local mock. It is not a deploy-ready credential or permission to connect customer systems.
The promise made in Track A — least privilege, human gates, a complete log — has to hold in identity providers, data stores, connectors, and production incidents. Entry requires experience implementing and debugging typed application code, asynchronous control flow, HTTP/auth boundaries, tests and test doubles, Git, persistent state and basic failure recovery. The entry diagnostic is to validate malformed JSON, implement a scoped mock read, write positive/negative assertions and explain an interrupted write. If those are unfamiliar, take a preparatory engineering route before B1; this is not a three-week beginner path. They do not assume a particular cloud or a shipped FusionWork API shape; examples are teaching models, and will be aligned to the product as it hardens.
Reading is not enough. An FDE who can explain a gate but has never built one, never sat in a discovery call, and never written a handover the owner can run, is not ready to deploy at a customer. Track B specifies that pathway:
- Lessons — the same short-module shape as the rest of the academy, with architecture, configuration, failure modes, and handover notes.
- Labs — every module in Volumes 1 to 5 ends in a lab in one starter repo, checked by tests the learner can run locally. Python is the working default, using Python 3.12 with type hints and dataclasses. Runtime support is pending verification. Install the released
pyproject.tomlpackage and verifieduv.lockwithuv sync --frozen; run cumulative checks such asuv run --offline pytest -m lab_b1_05. the lab repository will be public, containing fictional Lane & Hoy fixtures only, no real customer data and no secrets. Assessment-specific reviewer responses and variation seeds stay outside candidate-controlled public state. Installation may download packages; execution after setup is offline, using a deterministic mock agent, with no API keys, database, Docker or cloud requirement. The complete starter release remains separate from this curriculum. - Customer-facing practice — Volume 6 brings the FDE model forward from Track E and teaches discovery, scoping, saying no, and the weekly check-in.
- A capstone — the learner deploys an agent end to end for a mock Lane & Hoy brief: discovery, scope, build, evaluation, deploy, and handover.
- Sign-off — a deploy-ready rubric and monthly capstone presentation judged by FusionWork executives. Each slot is 1 hour: 30 minutes of presentation, 15 minutes of Q&A and 15 minutes of buffer.
- Shadowing and observed sign-off — supervised integration and incident exercises, then explicit senior post-shadow sign-off from FusionWork executives before a first lead engagement, which remains reviewed and may be supervised.
The engagement lifecycle
Volumes teach by topic. A real engagement runs in stages. Use this map to know which volume you need, what each stage should produce, and how you know it is done.
| Stage | What you produce | You can move on when | Where it is taught |
|---|---|---|---|
| 1. Understand, redesign and prove | Discovery notes, a redesign hypothesis, the real workflow, a scoped brief, success measures, a small eval set, a go-live conditions list | The owner agrees the scope and the measure, and a prototype passes the agreed checks | B1.01 owner brief; B6.02–B6.05; B4.01–B4.02; B7.03–B7.04 |
| 2. Build | Agent identity, scopes, gates, log, connectors — reusing what already exists | Gate and log tests pass, unhappy paths included | Volumes 1–3 |
| 3. Ground in knowledge | Sources with owners and refresh dates, a citation format, a real-question recall check | The agent cites what it used, and stale or conflicting entries are held for owner resolution | B3.05, B3.09, Track C |
| 4. Revisit the redesign | Who does what (people, agent, systems), checkpoints, exception paths | Every step is placed on the certainty × consequence grid, and costly steps are gated | B6.04, B3.07, B4.06 |
| 5. Operate | A quiet pilot, status, on-call, a weekly check-in, a handover pack | The owner can run it without you, and the four operating lenses are reviewed weekly | Volume 5, B6.08 |
Two deliverables on every engagement
The FDE delivers a customer-owned workflow and a short, de-identified method or asset at a stated level: used once, reusable, or standard. Reuse needs a rights/confidentiality check as well as removal of identifying detail; deleting names alone is insufficient. The appendix gives twelve methods, two companions and a short template. No customer source records belong in a public method.
“About three weeks plus capstone” is an unmeasured planning estimate for an existing engineer studying roughly 35–40 hours a week with the supplied scaffolding and timely feedback. The catalogue time boxes omit some reading, setup, help, review and rework; they are not a measured completion promise. A learner pilot must separately record those times, especially B1.05, B3.07, B3.09 and expanded B6 studios. Reserve reviewer/mentor capacity separately, including role-play, handover and remediation. Publish a measured range and a part-time route after that pilot. Sign-off depends on evidence, not elapsed days.
At B1.01 start an owner brief and a success scorecard: authorised AP sources, outcome, exclusions, Mei and backup, review capacity, hold/stop route, three baseline invoices, correct-match count, unsupported sends/payments/deletes = zero, and a target for time to a reviewed draft. Revisit discovery/redesign in B6.02–B6.05 while building, not only after Volume 5. B4 formalises and versions this scorecard before tuning. Keep the same growing dossier through the capstone.
The public lab specifications define checkpoints, supplied scaffolding, fault fixtures and acceptance. initial release scope is the eight B1 labs; B2–B7 stay specified for later delivery. The shared execution contract in the lab specifications is authoritative across the lessons: the complete envelope is hashed, policy alone mints checked tokens, scope and approval are independent, intent precedes dispatch, and outcomes are reconciled.
Capstone format
A monthly capstone presentation to a panel of FusionWork executives as judges. Each slot is one hour: 30 minutes of presentation, 15 minutes of Q&A and 15 minutes of buffer. Evidence collection, role-play and operator exercises happen separately from the presentation slot.
The assessed end of the programme. The learner deploys an agent for a mock Lane & Hoy brief, is reviewed against a deploy-ready rubric, and then completes supervised integration/incident work and shadowing, followed by a senior’s explicit decision before a first reviewed lead engagement.
B7.01 Quiet pilots, then package
Outcome: You can take two small, time-boxed pilots rather than one large programme, and package after the first useful result.
- Time to first useful result is the test.
- A second pilot is how you find out whether the method travels.
Studio B7.01: Propose a two-pilot plan for Lane & Hoy: AP first, then a tightly scoped support-triage follow-on, with a packaging step between them. (~45 min)
B7.02 Access is part of the critical path
Outcome: You can plan for access, processing authority, and data quality as if they were the build — because they often take longer.
- A four-day build with a three-week access wait is still a three-week-plus project.
- Tell the customer that, early, in calm language.
Practice note: Plan time for access and data-quality work, not only implementation. In one later case the build took days while access took weeks, and a data-quality issue held the result below the target.
Studio B7.02: Write the access plan for the capstone: who grants what, what you will do while you wait, and the sentence you will say to Mei when a grant is still outstanding on day four. (~45 min)
B7.03 The Lane & Hoy capstone brief
Outcome: You can read a customer brief, ask the questions it leaves open, and refuse to start building until the outcome, gates, and approver are named.
- The capstone changes from familiar AP to delivery exceptions, with an unseen constraint and owner reply controlled by the reviewer.
- Write back before you code.
Studio B7.03: Read the capstone brief, list the open questions, and send a one-page reply that either confirms the scope or recuts it. Do not start the build in this module. (~60 min)
B7.04 Capstone: discovery and scope
Outcome: You can run discovery against the Lane & Hoy packet, write the four-part handoff, and get a (mock) owner sign-off on the scope before you build.
- One workflow, one team, one time to first useful result test.
- Saying no, or recutting, is allowed and sometimes required.
Capstone phase 1: Produce a discovery note, a scoped brief, a role and scope map, and a written time to first useful result test, and obtain the reviewer-controlled mock owner decision after live discovery/scope role-play. The candidate cannot sign for the owner. (~half a day)
B7.05 Capstone: build, evaluate, and deploy
Outcome: You can implement the scoped workflow in the starter repo, pass the eval set and the gate tests, and promote it to a quiet pilot configuration.
- Sending and paying stay gated. Some candidates will keep the last mile off; that is a defensible choice if the brief allows it.
- Unhappy paths are tested, not only the demo path.
Capstone phase 2: Build, eval, red-team, and promote the workflow. The cumulative suite must pass, and the reviewer injects an unfamiliar partial-failure, expiry or evidence fault. Diagnose it and justify any refusal or rescope; evidence decides the result. (~two to three days)
B7.06 Capstone: handover and review
Outcome: You can leave a handover pack the owner and the next engineer can use, expire your temporary access, and sit a review of the whole path.
- The first page should be recognisable to Mei.
- Temporary admin access should already have expired before the review starts.
Capstone phase 3: Have another operator diagnose, freeze, revoke, reconcile and add a backup using only the handover pack. Submit their observed results, eval/red-team evidence and one de-identified asset with an honest level. Temporary access expires before review. Present at the monthly capstone review judged by FusionWork executives: a 1-hour slot with 30 minutes of presentation, 15 minutes of Q&A and 15 minutes of buffer. (~half a day for handover preparation/exercise, plus the presentation and any revision)
B7.07 Deploy-ready sign-off
Outcome: You can read the rubric, judge your own capstone against it, and understand what a reviewer will and will not pass.
- Identity, gates, and the log are must-pass. A strong demo does not offset a skipped control.
- "Deploy-ready" means ready to shadow, not ready to be left alone at a customer.
The rubric and the checklist are in the lab specifications, and are summarised at the end of this volume. FusionWork executives judge the capstone and score the assessment against the rubric, using studio and practical evidence gathered with technical reviewers. Evidence collection, role-play and operator exercises happen separately from the monthly 1-hour presentation slot; the slot does not lower the assessment bar. The learner also self-scores first; the gap is useful.
Studio B7.07: Score your capstone against the rubric, write one paragraph on the weakest must-pass area, and list the work you would still do before a real shadowing week. (~45 min)
B7.08 Shadowing before you lead
Outcome: You can explain how pairing on one real engagement works, what you may do, and what you must not do, before you lead your own.
- Shadowing happens off the site, on a live engagement, with a named senior FDE.
- The academy can only prepare you for it and ask you to write it up honestly afterwards.
The shadowing guide lives in the lab specifications. It is a brief for the learner and the senior, not a public case study of any customer.
Studio B7.08: Write your shadowing plan: what you want to observe, the limits you will keep, and the after-action format you will use. Do not name a real customer. (~30 min)
Sign-off rubric
A candidate is not deploy-ready if any must-pass row is below the pass bar. The lab specifications give area-specific Not yet / Developing / Pass anchors, non-waivable failures, reviewer evidence and calibration cases. All must-pass rows and minimum operational controls must Pass; other rows must be at least Developing with a remediation owner and date. All areas must Pass before FusionWork executives give senior post-shadow sign-off for a first lead.
| Area | Must-pass? | What a pass looks like |
|---|---|---|
| Discovery and scoping | Yes | Found the real workflow, recut or said no where needed, wrote a time to first useful result test the owner recognises. |
| Identity and least privilege | Yes | Agent has its own principal, scopes match the job, revocation was tested, no tenant-wide convenience grant. |
| Gates and irreversible actions | Yes | Send, pay, and delete are hard stops, bound to the complete execution envelope, with a named approver and a backup; no prompt-only rule, no timeout auto-approve. |
| Action log | Yes | Append-only, write-before-execute, answers what happened, who approved it, and whether those two things were the same. Personal data in the log is minimised. |
| Evaluation and safety | Yes | Eval set includes awkward cases; unhappy gate paths are tested; critical findings are fixed; only noncritical findings may have bounded, authorised risk acceptance. |
| Observability and operations | Minimum controls must Pass | Freeze, revoke, reconciliation and named on-call must work; richer reporting may be Developing. |
| Handover and reusable method | Yes | Short pack the owner and the next FDE can use; temporary access expired; backup approver can be added without the original engineer; one de-identified reusable asset with an honest level. |
| Process and value judgement | No, but required to be at least developing | Real workflow found from evidence; steps placed on the certainty × consequence grid; one sensible redesign; an honest value estimate that counts the hidden costs. |
| Customer-facing conduct | No, but required to be at least developing | Discovery notes, a scoped brief, a weekly check-in the owner could attend, and human review treated as part of the method. |
"Deploy-ready" signs the candidate off to shadow, not to lead unsupervised. FusionWork executives give the senior sign-off before a first lead; the first engagement remains reviewed and may be supervised.
Assessment levels: Not yet / Developing / Pass
The area-specific assessment levels and evidence requirements follow.
“Deploy-ready” means ready to shadow. Volume 1 alone has no such credential. Scores are Not yet / Developing / Pass. Discovery, identity, gates, action log, evaluation and handover are must-pass. Operations has mandatory minimum controls even if reporting is still Developing. Process/value and customer conduct must be at least Developing with a remediation owner, evidence target and due date. Before a first lead engagement every area must reach Pass, followed by explicit senior post-shadow sign-off from FusionWork executives. The first lead remains reviewed and may be supervised; FusionWork executives hold the judging and sign-off authority. Non-waivable failures: unapproved send/pay/delete; forged/replayed identity or token accepted; absent/revoked scope bypass; cross-workflow/unauthorised disclosure; missing durable intent; blind retry of unknown acceptance; ineffective freeze/revoke; critical ungrounded financial action; uncontrolled retention exposure; or fabricated owner/reviewer evidence. No name/date, attractive demo or aggregate score waives these. Only noncritical residuals can be accepted by the authorised customer owner and reviewer with rationale, compensating control and expiry. Expired acceptance blocks promotion.
Discovery and scoping
Threshold: Must Pass.
| Score | Observable anchor |
|---|---|
| Not yet | Builds before resolving authority or cannot find the changed constraint. |
| Developing | Finds some gaps but needs prompts to recut scope or make the success test measurable. |
| Pass | Independently discovers the constraint, negotiates a bounded outcome and obtains authentic owner agreement before build. |
Reviewer evidence: Recorded role-play, versioned brief/scorecard, reviewer-issued scope decision and refusal/rescope.
Identity and least privilege
Threshold: Must Pass.
| Score | Observable anchor |
|---|---|
| Not yet | Impersonates a person, accepts forged identity, or allows excessive/absent scope. |
| Developing | Separate identity exists but evidence for lifecycle, delegation or boundary checks is incomplete. |
| Pass | Distinct principals and acting-for authority, minimum resources/constraints, active-person checks and observed live revoke all hold. |
Reviewer evidence: Grant/decline map, forged-context and revoked-delegation tests, changed-directory diagnosis.
Gates and irreversible actions
Threshold: Must Pass.
| Score | Observable anchor |
|---|---|
| Not yet | Any mandatory action bypass, partial-envelope approval, forged/replayed token or timeout approval succeeds. |
| Developing | Happy path is correct but some state/race/backup/expiry evidence is missing. |
| Pass | Complete envelope, trusted token origin, expiry/withdrawal/replay checks and all send/pay/delete paths work under positive and adversarial cases. |
Reviewer evidence: Cumulative gate suite, attachment-only mutation, delete-target change, concurrent retry and reviewer variants.
Action log
Threshold: Must Pass.
| Score | Observable anchor |
|---|---|
| Not yet | An effect lacks prior durable intent or personal data is exposed/retained without controls. |
| Developing | Normal records exist but restart, unknown outcome or controlled expiry/correction is not yet demonstrated. |
| Pass | Intent precedes adapter entry, receipt/outcome reconcile after restart, access is scoped and retention/correction reaches relevant copies. |
Reviewer evidence: Boundary assertion, crash matrix, review query and expiry/restore evidence.
Evaluation and safety
Threshold: Must Pass.
| Score | Observable anchor |
|---|---|
| Not yet | Weak matcher/always-abstain solution passes, critical findings are waived or answers lack support. |
| Developing | Useful tests exist but labels, denominators, held-out variants or runner mutation evidence are incomplete. |
| Pass | Independent labels, useful coverage/grounding, versioned baseline, adversarial variants and failing broken-control runs support the release bar. |
Reviewer evidence: Eval manifest/results, independent owner disagreement record, mutation matrix and resolved critical findings.
Observability and operations
Threshold: Minimum controls must Pass.
| Score | Observable anchor |
|---|---|
| Not yet | Freeze/revoke/reconciliation fails, no accountable on-call exists or queued effects bypass stop. |
| Developing | Minimum freeze/revoke/recovery and on-call work; summary quality or workload reporting still needs assistance. |
| Pass | A different operator diagnoses a new fault, stops safely, reconciles and restores from versioned evidence; reporting is accurate and scoped. |
Reviewer evidence: Unfamiliar incident observation, queue/retry freeze tests, rota and recovery/summary evidence.
Handover and reusable method
Threshold: Must Pass.
| Score | Observable anchor |
|---|---|
| Not yet | Operator cannot use the pack, temporary privilege remains, or method exposes restricted data. |
| Developing | Pack looks complete but independent operator evidence, rights review or reuse limits are incomplete. |
| Pass | Operator executes required tasks without author coaching; temporary authority is gone; synthetic method has an honest level and reuse conditions. |
Reviewer evidence: Operator task record, config comparison, expiry test and method rights/de-identification check.
Process and value judgement
Threshold: At least Developing to shadow.
| Score | Observable anchor |
|---|---|
| Not yet | Claims benefit without baseline/denominator or automates the documented process despite contradictory evidence. |
| Developing | Maps the real process and a plausible redesign but cost/uncertainty or tail-delay reasoning needs help. |
| Pass | Uses central and tail evidence, states unknowns, tests redesign and counts exceptions, review, change and risk in net value. |
Reviewer evidence: B6.02 computed facts and B6.04 sensitivity/estimate with evidence and assumptions.
Customer-facing conduct
Threshold: At least Developing to shadow.
| Score | Observable anchor |
|---|---|
| Not yet | Promises unsupported capabilities, hides uncertainty or ignores the owner’s objection. |
| Developing | Communicates respectfully and records decisions but needs coaching on scope disagreement or next checks. |
| Pass | Conducts discovery, explains refusal and uncertainty, confirms owner understanding and leaves an actionable review decision. |
Reviewer evidence: Observed role-play and scope negotiation, operator briefing and weekly check-in decision.
Reviewer protocol, remediation and calibration
FusionWork executives judge the monthly capstone presentation. Each slot is 1 hour: 30 minutes of presentation, 15 minutes of Q&A and 15 minutes of buffer. Review the submitted evidence before the slot; the slot does not replace practical assessment or lower the rubric bar. Technical reviewers may assist with evidence collection and calibration; FusionWork executives own the assessment decision and senior sign-off before a first lead.
Separately from the presentation slot, reserve approximately 30 minutes to inspect the exact-version clean/cumulative/mutation reports, 30 minutes for discovery/scope role-play, 30 minutes for an unfamiliar failure and 30 minutes for operator handover, plus scoring and rework. These are capacity planning estimates, not measured timings; pilot reviewer agreement and workload separately from learner hours. Normal engineering tools, including AI help, are allowed; the candidate must explain and adapt the result.
Before review, verify authentic scope sign-off, current release fingerprint, expired temporary FDE access, self-score and fixture provenance. During review, observe positive and denied send/pay/delete; changed attachments; forged identity/token; expiry/replay; live revoke; crash/unknown-outcome recovery; freeze under queued/concurrent retries; data isolation and retention. Inspect durable intent at adapter entry. Then run the discovery, unfamiliar-fault and independent-operator tasks. Score every area with a cited observation, not a slide impression.
For a borderline result, a second reviewer checks the disputed evidence against these anchors and submits the findings to FusionWork executives for the decision. Record Not yet when mandatory evidence is absent. Give each gap an owner, corrective task, evidence target and review date; rerun affected cases plus cumulative invariants on the new version. Do not average a critical failure away. Store assessment records privately; publish only fictional teaching materials.
Calibration examples:
- Polished pack, but a changed attachment sends under an old approval: Not yet for gates and no readiness sign-off.
- All stop/revoke/reconciliation tests pass, but weekly review undercounts human review minutes: operations Developing; may be ready to shadow if all other thresholds hold, but not ready for first lead sign-off.
- The candidate refuses auto-send, obtains a reviewer-approved draft-only scope, detects an unknown receipt and enables another operator to reconcile it: relevant areas can Pass, even with fewer features.
- Perfect recall score with unsupported citations or universal abstention: evaluation Not yet, not a high-quality answer system.
- Learner attended shadowing but has no observed independent diagnosis or unresolved Developing areas: remains ready to shadow, not approved to lead.
Supervised integration and incident bridge
The public core labs use deterministic fixtures and require no model API keys or cloud account. Before real leadership, a named senior arranges separate authorised exercises outside those public labs: real IdP/sign-in and scoped grant/revoke; a provider sandbox connector with authentic errors/rate limits/unknown acceptance; genuine model/schema and indirect-injection behavior; reviewed multilingual/OCR examples; data-flow/transfer and retention/restore verification; and an incident drill with freeze, evidence preservation, customer privacy/incident escalation, recovery and notification decision routing. Customer/sandbox authorisation, least access, named owner/backup and stop routes are prerequisites; secrets and real data never enter public submissions. The senior records observed results and limits, not API keys or customer contents.
Shadowing and senior sign-off
Shadowing is a paired week (or the length of one engagement slice) on a real FusionWork customer, off this site. The learner observes discovery, access, a weekly check-in, and handover; may draft under review; and does not change production access, gates, or customer-facing commitments without the senior FDE. Afterwards they write a short after-action in a private assessment record. The supervising senior supplies observed evidence, and FusionWork executives explicitly give the senior post-shadow sign-off after all Developing areas are closed, with the first lead engagement's scope, supervision arrangements and executive review authority recorded. Attendance or an after-action alone gives no permission to lead. The academy publishes the guide, not the engagement.