[ Syllabus ]
All 26 weeks, in full
Every week: the mission, what happens in the FY28 case, the chapters you read, the documents you produce, the tools you use, the code lab, the agent lab, the business thread and the standard you're reviewed against on Friday.
- 26weeks
- 129deliverables you produce
- 152chapters, all read
- 47tools, your own accounts
No weeks match that filter.
Foundations and planning
You learn how a real company is wired and how its projects start, then plan all four projects to the standard a steering committee would sign. You write the business case, the charter, the RFP and the purchase request yourself. Python starts on day one.
You finish withAn integrated project management plan, all 18 components, defended live in week 8.
Week 1
How a company is wired
Understand the machine before you try to steer it: frameworks, structures, governance and GitLab's operating system.
5 deliverables
In FY28February 2027. FY28 begins. Four projects are about to collide inside one all-remote company of about 2,000 people. You meet the cast.
You produce
- Handbook pageYour personal README, merged into the cohort handbook by MR
- DocumentFramework map: where PMBOK 6, 7, 8, Scrum, Kanban and Lean each fit
- DocumentOrg analysis: structure, decision rights and where the PM sits for each of the four projects
- SlackLow-context introduction in #cohort: who you are, what you need, by when
- IssueYour first issue: a 26-week learning plan with milestones
You read
- IntroBefore You Begin: The portfolio you will follow: Ledger, Beacon, Run the Engine, Harbor
- Ch. 1What project management actually is: value not activity, the constraints you trade, three myths
- Ch. 2The map of frameworks: PMBOK 6 processes, PMBOK 7 principles, PMBOK 8, the agile family, tailoring
- Ch. 3How a company is wired: structures, PMOs, governance, EEFs and OPAs
- Ch. 4The GitLab operating system: CREDIT values, handbook-first, async, DRIs, iteration, TeamOps
Code lab
Python 1. Install Python, Jupyter and DuckDB. Variables, strings and lists; parse a messy task list exported from a tracker.
Agent lab
Install Claude Code or Codex. Have it scaffold your GitLab project (labels, issue templates, a README) through the API, then review every change before you accept it.
Business thread
How the company works, chapters 63–65. The company as a system, the corporate calendar, the executive office and the board. Exercise: Map the FY28 corporate calendar: closes, quarter ends, board dates, change locks.
Tools
ReviewMonday kickoff. Friday: three READMEs reviewed live for low-context writing.
Week 2
The workbench
Set up where the work lives, then learn to write the five things a PM writes every day.
6 deliverables
In FY28Ledger's program page goes up: status Build, health Needs attention, quick links to #proj-ledger, epic &812 and the steerco folder.
You produce
- Handbook pageProgram home page and document register, Exhibit 5.2 style
- BoardFour GitLab workspaces: scoped labels, boards, milestones and description templates
- IssueAn issue description: problem, proposal, out of scope, acceptance criteria, links
- Merge requestA merge request to the handbook: why, what changed, who must review
- Slack"hey chen can you check the workday thing" rewritten as a request someone can act on
- DocumentA live-doc meeting agenda with decisions and owners recorded in the doc
You read
- Ch. 5Documents, systems of record and the single source of truth; where each PMBOK artifact lives
- Ch. 6Issues, epics, labels, boards and milestones; setting up the four workspaces; hygiene
- Ch. 7The tool landscape: GitLab's actual stack, tools by phase, integrations, choosing tools
- Ch. 8Writing that moves work: requests, answers, issue and MR descriptions, SPADE, status, escalations, Slack
- Ch. 9Meetings worth having: whether to meet, live-doc agendas, the standard project meetings
Code lab
Python 2. Dictionaries and loops. Load a GitLab issues export (JSON) and count open work by label and assignee.
Agent lab
The agent drafts issue and MR descriptions from rough meeting notes. You rewrite them to the book's standard and log what it got wrong.
Business thread
How the company works, chapters 66–68. Strategy and business operations, FP&A, accounting, the close and SOX controls. Exercise: Trace one project dollar from the annual plan to a month-end accrual.
Tools
ReviewFriday: the cohort's rewritten Slack requests, judged by one test: could a newcomer act on it without asking a question?
Week 3
Where projects come from
Make the case for a project, charter it three different ways, and find every stakeholder before they find you.
7 deliverables
In FY28Feb–Mar 2027. Ledger enters intake and Ravi posts "BRD v0.9 ready for review" in #proj-ledger. Beacon starts problem validation with twelve customer interviews. Harbor's SOW-4471 sets a fixed price of $420,000.
You produce
- DocumentLedger business case with a benefits management plan
- IssueBeacon opportunity canvas as a GitLab issue
- DocumentLedger program charter, approved by the sponsor in a recorded comment
- EpicBeacon's epic description, written as its charter
- SpreadsheetStakeholder register and engagement assessment matrix (restricted)
- SpreadsheetRACI with exactly one A per row
- SlackA BRD review request: sections per reviewer, comment deadline, approval date
You read
- Ch. 10From strategy to project, portfolio choices, the business case, benefits management, the opportunity canvas, the SOW
- Ch. 11The charter: Ledger's classic charter, Beacon's epic-as-charter, Harbor's engagement charter, the working-group page
- Ch. 12Stakeholders: identify, analyze, register, engagement matrix, RACI, DACI and the DRI
Code lab
Python 3. Functions and CSV. Turn the stakeholder register into data and plot power against interest.
Agent lab
The agent drafts a charter from a one-paragraph brief. You red-line it against Template A.1 and the sponsor's questions.
Business thread
How the company works, chapters 69–71. Treasury, tax and billing; procurement and vendor management; legal, contracts and privacy. Exercise: Read a real-world contract stack: MSA, order form, SLA, DPA, security addendum.
Tools
ReviewFriday: three charters defended against the sponsor's questions. Standard: measurable success criteria and named decision rights.
Week 4
Logs, kickoff and the life cycle
Write down what you are assuming before it hurts you, start the team properly, and choose how each project will run.
6 deliverables
In FY28Spendwise's Okta SCIM support rests on its RFP answer alone. The team logs it as assumption A-01 with a sandbox test before go-live. The Finance close blackout becomes constraint C-01.
You produce
- IssueAssumption and constraint log as labeled issues (A-01, C-01)
- BoardRAID board: risks, actions, issues and decisions on one board
- DocumentHarbor internal handover: what the engagement manager sold and assumed
- SlidesKickoff deck and agenda
- Handbook pageTeam charter: working agreements written in a 40-minute session
- DocumentLife cycle decision for each of the four projects, with the reasoning
You read
- Ch. 13Assumptions, constraints, the assumption and decision logs, RAID on one board, keeping logs alive
- Ch. 14Kickoff in five steps: internal handover, align leaders, the kickoff, the team charter, make the team productive
- Ch. 15Choosing the life cycle: predictive, adaptive, hybrid; scaling; recording the choice
Code lab
Python 4. pandas DataFrames. Filter the RAID export by type, owner and age; flag items untouched for 14 days.
Agent lab
Give the agent a kickoff transcript; it proposes RAID entries. You flag every vague, duplicated or invented item.
Business thread
How the company works, chapters 72–74. People, talent and rewards; sales and revenue operations; marketing. Exercise: Find the quarter-end and quota calendar that will constrain Harbor's customer.
Tools
ReviewFriday: kickoff decks presented in five minutes each. Standard: every risk written as cause, event, effect, with an owner.
Week 5
Requirements, scope and estimates
Turn needs into requirements, requirements into scope, and scope into estimates with an honest basis.
6 deliverables
In FY28Ledger's BRD reaches v1.3 after 61 review comments. Beacon writes its first user stories. Harbor's scope is defined by counts: 8,000 repositories, 40 templates, 6 training sessions.
You produce
- DocumentRequirements management plan and BRD sections
- IssueUser stories with testable acceptance criteria
- SpreadsheetRequirements traceability matrix
- DocumentProject scope statement with exclusions and an acceptance process
- EpicWBS as an epic hierarchy, with a WBS dictionary entry per work package
- SpreadsheetThree-point estimates with a written basis of estimates
You read
- Ch. 16Requirements: types, the requirements management plan, collecting, writing, prioritizing, tracing (RTM), ready and done, spikes
- Ch. 17Scope: scope statement, WBS and dictionary, the baseline, the backlog, scope creep and gold plating
- Ch. 18Estimation: estimates vs targets vs commitments, the cone of uncertainty, techniques, basis of estimates, contingency
Code lab
Python 5. Group-by and roll-ups. Aggregate three-point (PERT) estimates up the WBS and show the range, not one number.
Agent lab
The agent proposes a WBS. You fix the decomposition against the 100% rule and record every correction.
Business thread
How the company works, chapters 75–77. Customer success, support and professional services; partners and channels; the engineering organization. Exercise: Follow a customer from signature to renewal and mark every handoff.
Tools
ReviewFriday: WBS walk-throughs. Standard: deliverable-oriented, 100% rule, every work package has an owner and acceptance criteria.
Week 6
Schedule, cost and quality
Build a critical-path schedule, a time-phased budget and a quality plan, then reconcile the budget with when money actually arrives.
5 deliverables
In FY28Ledger's Q3 spend plan is $40k above FP&A's allocation. Priya moves the Apex milestone-4 payment to hypercare exit in Q4 instead of moving any work. It takes two emails and a calendar check.
You produce
- SpreadsheetLedger schedule network (about 40 activities) with the critical path and float
- SpreadsheetCost baseline S-curve, contingency and management reserve
- SpreadsheetFunding requirements by fiscal quarter, reconciled with FP&A's allocation
- EmailThe two emails that move $40k of payments from Q3 to Q4 without moving work
- DocumentQuality management plan and definition of done
You read
- Ch. 19Schedule: activities, sequencing, the critical path method, resource optimization, compression, the baseline, adaptive scheduling
- Ch. 20Cost and budget: estimates, the cost baseline S-curve, contingency vs management reserve, funding-limit reconciliation
- Ch. 21Quality: principles, the quality plan, building quality in, the seven basic tools, quality in a fixed-price engagement
Code lab
Python 6. Graphs. Compute Ledger's critical path in code and check it against your spreadsheet model.
Agent lab
The agent computes the critical path and draws the S-curve. You check both by hand before either goes in the plan.
Business thread
How the company works, chapters 78–80. Infrastructure, SRE and platform teams; security and trust; IT and enterprise applications. Exercise: Write the security and privacy review steps into Ledger's schedule as real durations.
Tools
ReviewFriday: schedules stress-tested live (what happens if card feeds slip five days?). Standard: no date without a network behind it.
Week 7
Resources, communications and risk
Plan the people you actually have, the messages each audience needs, and the risks that will decide your dates.
5 deliverables
In FY28Priya runs 20,000 Monte Carlo simulations of Ledger's schedule. The 82-day plan is met in only 37% of runs; P80 is 87 days. External commitment: 4 October, with a fallback to 18 October.
You produce
- SpreadsheetCapacity plan for the Beacon and Run the Engine team, including unplanned work
- DocumentCommunications management plan, Harbor-style, across two organizations
- IssueRisk register as issues, each with cause, event, effect, owner, trigger and fallback
- DocumentDecision tree for the SCIM failure: wait, build a script, or slip go-live
- PythonMonte Carlo schedule simulation with P50, P80 and P95 dates
You read
- Ch. 22Resources and capacity: people are not units, acquiring resources, capacity planning, collisions, the utilization trap
- Ch. 23Communications: push, pull, interactive; the channels formula; Harbor's plan; the release post
- Ch. 24Risk: identification, qualitative and quantitative analysis, EMV, decision trees, Monte Carlo, responses, GitLab's StORM
Code lab
Python 7. Simulation. Re-run Ledger's Monte Carlo (20,000 runs, beta-PERT) and reproduce P50 ≈ 83 days and P80 ≈ 87 days.
Agent lab
The agent writes the simulation. You verify the distributions, the merge points and the random seed before trusting a single percentile.
Business thread
How the company works, chapters 81–83. The data function; risk, compliance and internal audit; corporate development and M&A integration. Exercise: Place your top five project risks inside the company's enterprise risk register.
Tools
ReviewFriday: risk registers challenged by a 'red team' of classmates. Standard: every high risk has a trigger and a fallback.
Week 8
Procurement and the integrated plan
Buy a $1.14M software platform the way a public company does, then pull every plan into one.
Phase review7 deliverables
In FY28Apr–Jun 2027. Ledger's RFP scores Spendwise 4.10, Expensa 3.90, TripTally 3.50. Then five weeks from 'yes' to signature: sanctions, security, privacy, legal, negotiation, FP&A, Coupa approvals, DocuSign, PO released 4 June.
You produce
- DocumentProcurement statement of work (RFP Annex B)
- SpreadsheetSource selection criteria with weights agreed before proposals are opened
- SpreadsheetIndependent cost estimate vs three partner proposals
- EmailBidder-conference Q&A published to every bidder in writing
- Purchase requestPurchase request: Zip intake, review path and Coupa approval chain (FP&A → VP IT → CFO)
- DocumentContract term sheet: liability cap, SLA exhibit, DPA, renewal cap
- Handbook pageIntegrated project management plan: where all 18 components live
You read
- Ch. 25Procurement: make or buy, contract types, RFP/RFQ/RFI, procurement SOW, source selection, independent cost estimates, GitLab's process, the seller's side
- Ch. 26Integration: the project management plan, change and configuration management plans, the performance measurement baseline
Code lab
Python 8. Sensitivity. Re-weight the vendor evaluation and find the weights at which Spendwise stops winning.
Agent lab
The agent compares the MSA, SLA and DPA and builds a term sheet. You verify every clause it cites against the source text.
Business thread
How the company works, chapters 84–86. Investor relations and communications, operations and workplace, the PMO across the company. Exercise: Run the cross-functional intake checklist on your capstone project.
Tools
ReviewPhase 1 review. You present and defend your integrated plan live. Pass, or revise and re-present within a week.
Execution and control
The projects go live and things go wrong on schedule: a vendor's connector fails, a production incident takes 12% of page views, a bank buys a subsidiary. You run sprints and flow, report honestly, control change and close. SQL starts here.
You finish withFour closed projects with status reports, an earned value history, a change order, an incident review and a retrospective.
Week 9
The operating rhythm and Scrum
Run the work week to week without micromanaging, then run Scrum properly inside a fixed-price contract.
Mini Career Day5 deliverables
In FY28Harbor's iteration 5, end to end: planning, daily coordination, review with Meridian Bank and a retrospective. Mini Career Day this week.
You produce
- IssueIssue log with owners, next actions and links to the risks that predicted each issue
- BoardIteration board with weights and a sprint goal
- DocumentSprint review notes and a retrospective with three actions
- SlackA daily async standup thread via Geekbot
- DocumentA one-page memo on why velocity is not a performance metric
You read
- Ch. 27The PM's operating rhythm: directing without micromanaging, the issue log, unblocking, dependencies, project knowledge
- Ch. 28Scrum done properly: accountabilities, events, artifacts, Harbor's iteration 5, velocity and its abuse, anti-patterns
Code lab
SQL 1. SELECT, WHERE and ORDER BY in DuckDB against an issues table. Velocity per iteration and its spread.
Agent lab
The agent summarizes a sprint-review recording transcript. You check every claim against the recording.
Business thread
Strategy, chapters 87–90. What strategy is, industry analysis and the Five Forces, competitive advantage, sustaining advantage. Exercise: Run the Five Forces on Larkspur, a $140M freight-software company, and rework its strategy statement.
Tools
ReviewMini Career Day #1: employers meet the cohort and see integrated plans.
Week 10
Flow, Lean and engineering practice
Manage work that never ends, and understand exactly how a change travels from idea to production.
5 deliverables
In FY28Run the Engine: bugs arrive daily with severity SLOs, security fixes override features, and a feature flag rolls out to 1%, then 10%, then 100% of groups.
You produce
- DocumentTriage policy with severity SLOs
- BoardKanban board with WIP limits and explicit policies
- DashboardFlow metrics: cycle time percentiles, throughput, a cumulative flow diagram you can read aloud
- DocumentValue stream map showing where work waits
- IssueFeature-flag rollout issue with steps, metrics to watch and the rollback command
You read
- Ch. 29Flow: why flow not sprints for run-the-engine work, Kanban, flow metrics, triage, cumulative flow diagrams, Lean
- Ch. 30Engineering practices PMs must understand: MRs and review, CI/CD, environments, feature flags, the test pyramid, tech debt, DORA
Code lab
SQL 2. GROUP BY and COUNT(DISTINCT). Cycle time percentiles and weekly throughput from issue history.
Agent lab
The agent analyses a flow export and claims a bottleneck. You confirm or refute it with your own SQL.
Business thread
Strategy, chapters 91–94. Corporate, global and technology strategy; executing strategy through alignment and ambidexterity. Exercise: Decide whether Larkspur should build, buy or partner for route optimization.
Tools
ReviewFriday: cumulative flow diagrams read aloud. Standard: name the bottleneck and the policy that would fix it.
Week 11
Teams, leadership and conflict
Lead people you don't manage, across time zones, through disagreement.
4 deliverables
In FY28An Apex consultant is replaced mid-project. Field Operations has to absorb a mid-quarter go-live. Meridian's platform lead challenges Harbor's forecast in front of the customer.
You produce
- DocumentTeam health diagnosis using Tuckman and Lencioni
- DocumentOne-to-one coaching notes with agreed actions
- DocumentNegotiation prep sheet: interests, BATNA, ZOPA, concessions
- SlackA hard message, written for a global team with tone made explicit
You read
- Ch. 31Teams: Tuckman, Lencioni's five dysfunctions, psychological safety, trust when the team never meets, cultures
- Ch. 32Leadership and motivation: styles, power, motivation, emotional intelligence, coaching, delegation, leading without a matrix
- Ch. 33Conflict, negotiation and decisions: five modes, negotiation, decision methods, hard conversations
Code lab
SQL 3. JOINs across issues, people and iterations. Find who is overloaded, and learn the fan-out join trap.
Agent lab
The agent plays a difficult stakeholder in a live role-play. You debrief what worked using the book's conflict modes.
Business thread
Financial accounting, chapters 95–97. Accounting foundations and double entry, the three financial statements, revenue recognition. Exercise: Post a software bundle through the five revenue-recognition steps.
Tools
ReviewFriday: live role-plays (conflict, coaching, negotiation), scored by classmates against a rubric.
Week 12
Stakeholders, steering committees and vendors
Keep executives engaged between meetings and hold vendors to the contract without poisoning the relationship.
6 deliverables
In FY28Steering committee #4, 26 August. The vendor scorecard rates Spendwise 3.0 and Apex 4.0. Owen raises an SLA credit claim for the late SCIM fix.
You produce
- SlidesSteering committee pack with the decisions needed on page one
- DocumentEnablement and change management plan for 2,000 employees
- SpreadsheetVendor milestone and payment tracker that keeps acceptance separate from payment (SOX)
- SpreadsheetVendor performance scorecard with evidence for every score
- EmailSLA credit claim under the SLA exhibit
- EmailAn escalation to the sponsor: facts, what you've done, the ask, the options
You read
- Ch. 34Stakeholder engagement: the steering committee done well, executives between meetings, leading change, customers and users
- Ch. 35Vendors: control procurements, Ledger's vendor landscape, scorecards, SLAs and credits, claims and disputes, transition
Code lab
SQL 4. Window functions. Vendor ticket response times against the SLA, median and p90 by month.
Agent lab
The agent drafts the steering committee pack. You cut it to the three decisions that matter.
Business thread
Financial accounting, chapters 98–100. Assets, liabilities and equity, fair value, investments and comprehensive income. Exercise: Decide what Ledger's spend capitalizes and what it expenses.
Tools
ReviewFriday: escalations reviewed. Standard: short, factual, an ask and options, no blame.
Week 13
Measuring progress honestly
Report status people believe, forecast with the right model, and run earned value on real numbers.
6 deliverables
In FY28Ledger at the end of August: PV $370k, EV $352k, AC $363k. CPI 0.97, SPI 0.95. One control account, integrations, explains half the variance.
You produce
- DocumentStatus definitions with objective RAG triggers
- SpreadsheetMilestone trend analysis
- DocumentLedger monthly work performance report
- EpicBeacon weekly async status note on the epic
- SpreadsheetEVM by control account with EAC by four methods and earned schedule
- DashboardA Tableau Public dashboard of CPI and SPI over the year
You read
- Ch. 36Measuring progress: status that people believe, milestone trend analysis, forecasting models, four status reports
- Ch. 37Earned value: PV, EV, AC, variances and indices, EAC methods, TCPI, re-baselining, earned schedule, AgileEVM
Code lab
Python + SQL. Earned value from a cost ledger: CPI, SPI, EAC, TCPI and earned schedule, month by month.
Agent lab
The agent computes EVM. You check every figure, and the method it chose for EAC.
Business thread
Accounting and finance, chapters 101–103. Financial statement analysis and managerial accounting, accounting to valuation, the time value of money. Exercise: Analyze Larkspur's profitability and unit economics from its statements.
Tools
ReviewFriday: status reports read by a mock executive. Standard: the first two lines carry the message.
Week 14
Change, incidents and quality
Control change four different ways, run an incident minute by minute, and find root causes without blame.
7 deliverables
In FY28INC-2219: a feature flag reaches 100% of GitLab.com at 13:40 UTC; epic boards time out; mitigated at 14:40. Meanwhile CO-1 adds 2,000 repositories and $96,000 to Harbor.
You produce
- IssueChange request with impact analysis, through the CCB
- DocumentChange order CO-1: scope, estimate basis, price, payment schedule, approvals
- IssueCommitment change request on an interlock epic
- IssueProduction change request classified C1 to C4
- SlackIncident channel: declaration, roles, status updates
- DocumentBlameless review: timeline, five whys, Swiss cheese, SMART corrective actions
- DocumentUAT sign-off and acceptance record
You read
- Ch. 38Change control: Ledger's CCB, Harbor's change orders, Work at Risk and Work Exceptions, Beacon's commitment change request, production change management
- Ch. 39Issues, incidents and root causes: GitLab's incident lifecycle, INC-2219 minute by minute, blameless review, five whys, Swiss cheese, corrective actions
- Ch. 40Validating scope and controlling quality: two gates, in order
Code lab
SQL. Rebuild an incident timeline from event logs; compute time to detect, mitigate and resolve; a Pareto of causes.
Agent lab
The agent drafts a postmortem from a raw timeline. You remove blame, add the missing facts and test each corrective action.
Business thread
Finance, chapters 104–107. NPV and capital budgeting, bonds and interest rates, risk and return, the cost of capital. Exercise: Compute Larkspur's WACC and decide which projects clear the hurdle rate.
Tools
ReviewFriday: a live incident simulation. You are the incident lead for 30 minutes.
Week 15
Closing and learning
Close projects, phases and contracts properly, and run retrospectives that actually change what happens next.
Phase review6 deliverables
In FY28Nov 2027–Jan 2028. Ledger leaves hypercare; Coupa POs close. Harbor closes with milestone acceptance, a customer retrospective and a CSAT survey.
You produce
- DocumentTransition to operations and hypercare exit
- DocumentFinal report against objectives and the business case
- Purchase requestProcurement closure: PO closed, accruals cleared, final invoice matched
- SurveyCustomer CSAT survey
- IssueAsync retrospective issue with owners for every action
- SpreadsheetLessons learned register entries that change a template
You read
- Ch. 41Closing: procurements, Ledger from hypercare to operations, the final report, OPAs, Harbor's close, when a product 'closes'
- Ch. 42Retrospectives and lessons learned: structure, a GitLab group retrospective, retros with customers, the lessons register
Code lab
Python. Burn-up charts across the year and a data-backed retrospective.
Agent lab
The agent clusters 80 retro comments into themes. You challenge every grouping before it reaches the team.
Business thread
Finance, chapters 108–111. Capital structure and payout, options and real options, M&A, corporate governance. Exercise: Value the option to expand Beacon if adoption beats target.
Tools
ReviewPhase 2 review: closure report and a retrospective you facilitate live.
The product
You move from running projects to deciding which work is worth doing, following Beacon from twelve customer interviews to GA: vision, OKRs, discovery, prioritization, specs, design, metrics, experiments, pricing, launch and AI features.
You finish withA product dossier: vision, OKRs, opportunity tree, scored backlog, roadmap, spec, metrics tree, experiment and an AI evaluation plan.
Week 16
The portfolio year and the product manager's job
Step back to see the whole year, then step into the product manager's chair.
3 deliverables
In FY28The four projects month by month: who resolved each collision, the year-end scorecards, and what changes for FY29.
You produce
- DocumentPortfolio collision analysis: four collisions and who resolved them
- DocumentYour first-90-days plan for a real role
- DocumentA comparison of the project, program and product manager roles at one company
You read
- Ch. 43A year in the portfolio: month by month, collisions, portfolio vs program vs project, scorecards
- Ch. 44Your practice: ethics, the many jobs called PM, the exam vs reality, your first 90 days
- Ch. 45The product manager's job: the product trio, GitLab's product principles, the product operating model
Code lab
SQL. Join four projects' metrics into one portfolio scorecard.
Agent lab
Start your personal AI playbook: the prompts, checks and failure cases that held up over 15 weeks.
Business thread
Marketing, chapters 112–114. Marketing strategy and competitive dynamics, how customers decide, research, segmentation and targeting. Exercise: Segment Ardent Outdoor's hiking customers and choose a target.
Tools
ReviewFriday: 90-day plans reviewed by a hiring manager on the panel.
Week 17
Vision, OKRs and discovery
Decide where the product is going and learn what customers actually need before anyone writes code.
Mini Career Day5 deliverables
In FY28Beacon: "when will it be done?" Twelve customer interviews, an opportunity canvas and a jobs-to-be-done map. Mini Career Day this week.
You produce
- DocumentPositioning statement
- IssueQuarterly OKRs as GitLab work items
- DocumentInterview discussion guide and two interview snapshots
- DocumentOpportunity solution tree
- DocumentAssumption map and test plan
You read
- Ch. 46Vision, strategy and positioning: the market, canvases, product-market fit
- Ch. 47Goals, OKRs and outcomes: committed vs aspirational, alignment without a rigid cascade
- Ch. 48Continuous discovery: methods, interviewing, JTBD, personas, opportunity solution trees, assumption tests, MVC
Code lab
SQL. Weekly active forecast groups with COUNT(DISTINCT), broken down by plan tier, excluding internal namespaces.
Agent lab
The agent synthesizes mock interview notes. You flag leading questions and claims the notes don't support.
Business thread
Marketing, chapters 115–117. Positioning and brand management, product lines and new products, pricing. Exercise: Price Ardent Outdoor's new tent line with value-based pricing.
Tools
ReviewMini Career Day #2: employers meet the cohort after sprints, EVM, an incident and a closed project.
Week 18
Prioritization, roadmaps and specs
Choose what not to build, show direction without false precision, and turn a problem into a shippable increment.
4 deliverables
In FY28Beacon's RICE score, a story map and a roadmap change note: forecasts on roadmaps move from Now to Next.
You produce
- SpreadsheetPrioritization worksheet with RICE and WSJF
- DocumentNow/next/later roadmap and a roadmap change note
- IssueFeature proposal issue sliced into minimal valuable changes
- SlackA 'no' to a big customer's request, written so the account team can use it
You read
- Ch. 49Prioritization: scoring, Kano, cost of delay and WSJF, MoSCoW, balancing bugs and features, saying no
- Ch. 50Roadmaps: formats, building one, one roadmap several views, changing the roadmap
- Ch. 51From problem to shippable increment: spec formats, quality attributes, design handoff, working with engineering
Code lab
Python. RICE and WSJF scoring with a sensitivity test on every guess.
Agent lab
The agent scores the backlog. You challenge the assumption behind each score.
Business thread
Marketing, chapters 118–120. Marketing communications and digital marketing, distribution channels, designing the sales force. Exercise: Design the channel strategy for Larkspur's EU expansion.
Tools
ReviewFriday: roadmaps presented to a skeptical sales leader.
Week 19
Design, metrics and experiments
Judge design with evidence, measure what matters, and run experiments that can actually fail.
4 deliverables
In FY28Beacon's forecast callout experiment and its dashboard, where accuracy improved for every type and stayed flat overall (Simpson's paradox).
You produce
- DocumentHeuristic review and accessibility check of a real flow
- DocumentNorth Star and metrics tree with metric definitions
- DashboardDashboard specification with questions as chart titles
- DocumentExperiment brief and results write-up
You read
- Ch. 52Design and UX for PMs: journeys, heuristics, discoverability, IA, content, accessibility, giving design feedback
- Ch. 53Product metrics: North Star, frameworks, core analyses, instrumentation, how analytics mislead
- Ch. 54Experimentation: design, the statistics you need, threats to validity, a worked experiment, small samples
Code lab
SQL and Python. A funnel and a cohort retention table with CTEs; an A/B analysis with confidence intervals; a Simpson's paradox check.
Agent lab
The agent analyses the experiment. You check the statistics and the denominator before anyone ships.
Business thread
Marketing and operations, chapters 121–123. Customer lifetime value and centricity, global marketing, operations strategy. Exercise: Compute customer lifetime value for Larkspur's segments.
Tools
ReviewFriday: experiment write-ups. Standard: a pre-registered decision rule.
Week 20
Business models, launch and AI products
Price and package a product, launch it in tiers, and build an AI feature you can evaluate and defend.
5 deliverables
In FY28Beacon's release post for GA in 20.4, a deprecation notice, and an evaluation plan for an AI feature.
You produce
- DocumentPricing or tier decision record
- DocumentLaunch plan and readiness checklist
- Merge requestRelease post as a merge request
- DocumentDeprecation notice
- DocumentAI feature evaluation plan with failure modes and a 20-case eval set
You read
- Ch. 55Business models, unit economics, pricing, packaging, product-led growth, pricing AI
- Ch. 56Go-to-market and launch: tiers, release stages, the launch plan, the release post, deprecations
- Ch. 57Growing, maturing and retiring products
- Ch. 58Building AI products: when to use AI, assistants to agents, building blocks, evaluation, cost, latency and quality
Code lab
Python. A unit-economics model (CAC, LTV, payback) you can defend line by line.
Agent lab
Build a small AI feature with your agent, write 20 evaluation cases, score the outputs against a rubric and report where it fails.
Business thread
Operations, chapters 124–127. Process design, process analysis and Little's law, waiting lines, quality and statistical process control. Exercise: Apply Little's law to Harbor's migration pipeline.
Tools
ReviewFriday: launch plans and AI evaluation results.
Week 21
Enterprise products and data-informed decisions
Work with the field, handle the big customer's request, and write decisions that say what would change your mind.
Phase review3 deliverables
In FY28Beacon's decision memo: scope-growth modeling before roadmap forecasts, with what would change our mind.
You produce
- DocumentEnterprise readiness checklist
- DocumentData-informed decision memo with 'what would change our mind'
- DocumentThree casebook decisions, argued and defended
You read
- Ch. 59Enterprise product management: how enterprises buy, readiness, deployment models, the field, advisory boards, escalations
- Ch. 60Data skills: data structure, enough SQL to be dangerous, statistics, spreadsheets, dashboards, checking data
- Ch. 61The craft and career of product management
- Ch. 62Product decisions: a casebook of twelve hard calls
Code lab
Data exercise: question to SQL to Python to a Tableau Public chart to a decision memo.
Agent lab
The agent writes the SQL. You verify the grain of every table and the result against a number you already trust.
Business thread
Operations, chapters 128–131. Forecasting, inventory, supply chain management, strategic sourcing. Exercise: Forecast Ardent Outdoor's demand and set its order quantities.
Tools
ReviewPhase 3 review: a casebook decision presented and defended, plus the data exercise.
Tools in practice
You rebuild your capstone in the systems employers actually run: GitLab in depth, Jira and Confluence, the Microsoft stack, spreadsheets that can be trusted, and the enterprise systems behind money, people and incidents. Then you automate the boring parts.
You finish withYour capstone in GitLab and Jira, a Microsoft Project schedule, a trusted workbook, a dashboard and one working automation.
Week 22
GitLab in depth, then Jira and Confluence
Set up a real project in GitLab step by step, then do it again in Jira so you can work in either shop on day one.
4 deliverables
In FY28The label and template cookbook: scoped labels, description templates, boards and saved views, and a triage policy written as code.
You produce
- BoardYour capstone in GitLab with a full label taxonomy and templates
- Merge requestTriage policy as code, merged by MR
- BoardYour capstone in Jira: epics, board, JQL filters, dashboard, an automation rule
- Handbook pageA Confluence project page and decision log
You read
- Ch. 145GitLab for project managers: work items, labels, milestones and iterations, views, collaboration, analytics, the handbook, Service Desk, Duo, recipes
- Ch. 146Jira and Confluence: vocabulary, JQL, dashboards, automation, Product Discovery, Service Management, Confluence, Rovo, recipes
- IntroApp. E: GitLab label and template cookbook
Code lab
Python. Pull issues through the GitLab API and Jira JQL and produce the same RAID report from both.
Agent lab
The agent builds a triage automation as a merge request. You review it like a maintainer would.
Business thread
Leadership and organizational behavior, chapters 132–135. Designing effective teams, leading global teams, decision making and judgment, negotiation. Exercise: Redesign a failing global team using the book's structure, process, language and identity levers.
Tools
ReviewFriday: tool tours. Five minutes to show your capstone in each tool.
Week 23
The Microsoft stack and spreadsheets you can trust
Schedule properly in Microsoft Project, build workbooks that survive an audit, and serve the people who live in Excel.
4 deliverables
In FY28A Microsoft-stack project, put together: Project, Planner, Teams, SharePoint Lists, Excel and Power BI.
You produce
- SpreadsheetMicrosoft Project schedule with baseline and resource leveling
- DashboardExcel or Power BI status dashboard
- SpreadsheetA trusted EVM workbook: inputs, calculations and outputs separated, every formula checked
- SpreadsheetSmartsheet tracker for a Smartsheet-first client
You read
- Ch. 147The Microsoft stack: Project desktop, Planner, Teams, SharePoint and Lists, Loop, Excel and Power BI, Azure Boards, Copilot
- Ch. 148Spreadsheets and Smartsheet: building PM workbooks that can be trusted, the standard workbooks, AI in spreadsheets
Code lab
Python. Generate a formatted Excel workbook from pandas for stakeholders who live in Excel.
Agent lab
The agent writes workbook formulas. You break them with edge cases before anyone else does.
Business thread
Leadership and entrepreneurship, chapters 136–139. Leading organizational change, becoming a manager, entrepreneurial opportunities, hypothesis-driven entrepreneurship. Exercise: Write Plumbline's hypotheses and design the cheapest test for each.
Tools
ReviewFriday: workbooks audited by a classmate. Standard: someone else can find any number's source in 30 seconds.
Week 24
Communication, enterprise systems and automation
Set up a project's whole communication system, understand every enterprise system around it, and automate the repetitive parts.
4 deliverables
In FY28The systems of record behind every project: Kantata for hours, Zip and Coupa for money, DocuSign for contracts, Salesforce and Gainsight for customers, incident.io and PagerDuty for outages, Tableau for answers.
You produce
- DocumentA communication setup for a new project: channels, standups, cadence, status page
- DocumentSystems-of-record map for your capstone
- DocumentTool selection memo for a hypothetical employer
- PythonAn automation that drafts the weekly status from GitLab and posts it to Slack after a human approves it
You read
- Ch. 149Documents, whiteboards, design and research tools: Google Workspace, Notion, Mural, FigJam, Figma, Dovetail, Loom
- Ch. 150Communication tools: Slack, Zoom, Geekbot, email and calendars, status pages
- Ch. 151Enterprise systems: PSA, PPM, ITSM, incident response, procurement, CRM, BI, finance and HR systems
- Ch. 152Other trackers compared, choosing, migrating between tools
Code lab
Python. GitLab API plus a Slack webhook: your weekly status draft, generated, reviewed and then posted.
Agent lab
The agent builds the end-to-end automation. You document exactly where a person checks its output.
Business thread
Entrepreneurship and innovation, chapters 140–144. Entrepreneurial selling, business plans and pitches, financing ventures, scaling, breakthrough innovation. Exercise: Pitch Plumbline's seed round in five minutes.
Tools
ReviewFriday: automations demoed live, including what happens when they fail.
Capstone and Career Day
No new reading. You assemble a public portfolio, sit a timed 160-question practice exam, rehearse interviews and present your work to employers.
You finish withA finished portfolio, interview practice and Career Day.
Week 25
Portfolio and exam practice
Assemble everything into a portfolio employers can open, and prove you can pass the exam.
4 deliverables
In FY28Your FY28: every document, decision, query and automation you produced, organized the way a hiring manager reads.
You produce
- Handbook pagePublic portfolio site on GitLab Pages
- DocumentTimed 160-question practice exam and your error log
- DocumentCAPM or PMP study plan
- PythonData appendix: your best notebook and queries, cleaned up and explained
You read
- IntroApp. A: Template library: 34 templates
- IntroApp. B–D: Formula sheet, PMBOK 6 process map, agile quick reference
- IntroApp. F: Document atlas
- IntroApp. I: 160 practice questions with explained answers
Code lab
Clean and document your three best notebooks and queries.
Agent lab
Mock interview with your agent as the interviewer; review the transcript and fix weak answers.
Business thread
Integration. No new reading. Exercise: Pick the three business ideas you would bring to your first job.
Tools
ReviewFriday: portfolio reviews by working PMs.
Week 26
Interviews and Career Day
Show employers what you can do.
Career Day3 deliverables
In FY28Career Day.
You produce
- DocumentLive mock interview with written feedback
- SlidesTen-minute portfolio presentation
- DocumentYour personal AI playbook, final version
You read
- IntroApp. G: Glossary, for interview preparation
Code lab
Walk an interviewer through one notebook and one query.
Agent lab
Finalize your AI playbook and include it in your portfolio.
Business thread
Career. Interview stories from the business thread. Exercise: Explain one business idea to a hiring manager in two minutes.
Tools
ReviewCareer Day: online, with every employer in our network.
Chapter numbers refer to Projects, For Real. The syllabus may change between cohorts as we improve it; the shape won't.
[ Operator Program · Cohort 1 ]
Your first real fiscal year starts here
Free, live and online. Applying takes about ten minutes, and you'll hear from us within 7 days.
The whole program is free. No tuition, no deposit, no income-share agreement, nothing owed if you leave.