The 3-Week LBO
Sources and uses, the full debt schedule, and returns, with attention to the covenant and capital-structure detail AI routinely gets wrong. Generic models miss the waterfall; you will not.
Forge · Modeling Sprints
Three weeks. One finished model. Proof you can build.
LBO, M&A, DCF: structured sprints where you build the model cell by cell, then layer AI to go faster. You leave with an artifact, not a certificate of attendance.
Why it matters
A finished model you can defend beats ten courses you completed.
Anyone can now generate a model that looks right. What separates the analyst who gets hired and kept is the one who can build it by hand, explain every assumption, and catch where a confident tool went quietly wrong. Sprints forge that, and hand you the proof: a working file on your own drive you can open, walk through, and demo in an interview.
We run them as three-week sprints because real deal teams work to deadlines, and deadlines forge focus. A brief, a clear deliverable, checkpoints, and a graded project. Plenty of people start an LBO. The sprint is built so you ship it.
Source: Knowledge / AI Capability Assessment 2026
/A portfolio that proves judgment.
A finished LBO you can defend is evidence, and evidence is what moves a hiring conversation.
/Build by hand, then accelerate.
Learn the mechanics manually first, so when AI drafts the scaffold you immediately spot what is wrong with it.
/Deadlines that forge focus.
Three weeks, a clear deliverable, checkpoints, and a graded project: the rhythm of a real deal team.
/Current by design.
Cases reflect what is moving now: private credit structures, carve-out financials, and AI-infrastructure project finance.
What's inside
Three sprints, one shape: brief, build, checkpoint, graded project.
The 3-Week M&A
Accretion and dilution, synergies, and the deal narrative, which sits squarely in human-required territory. You model the math and tell the story the math supports.
The 3-Week DCF
AI can scaffold a DCF in about 10 minutes, which is exactly why you build it yourself first, then learn to audit the AI version: the circular reference, the unsupported assumption, the drifted figure.
Proof and credibility
Claude for Excel reads the entire workbook before it responds and can scaffold a full DCF in about 10 minutes. Sprints teach you to use that speed and to pressure-test every assumption it makes, because the leading models still hallucinate figures, break on circular references, and stay blind to normalization. Mathematically consistent is not the same as correct.
Source: Knowledge / AI Capability Assessment 2026
Straight answers
Before you ask.
If AI builds it in 10 minutes, why spend three weeks?
Because the 10-minute draft is wrong in ways only a trained modeler catches, and a draft you cannot audit is a liability. The sprint forges the catcher.
Are sprints free?
Core sprints are part of the free student tier. Advanced and professional tracks may sit behind the subscription. Students learn free because professionals pay.
I have never built a full model, is this too advanced?
No. Sprints start from a Foundations-level baseline and scaffold up week by week. Building your first complete model is the point of being here.
Will I actually finish?
The three-week structure, the checkpoints, and a matched study pod are all designed so that you do. Finishing is the deliverable.
