Executive / Asset Mgmt
Executive / Asset Mgmt Track
See the whole portfolio at a glance, pressure-test a deal before IC, and turn raw diligence into board-ready narrative — with every figure traced.
Sam Simonian · CEO / Acquisitions
Asset Managers
Leadership
Anyone prepping IC or board materials
You're on this track if
- You make portfolio-level decisions and want the numbers synthesized, not raw
- You sit on IC and want assumptions stress-tested before the meeting
- You write or review IC memos, board materials, and investor updates
- You want AI leverage on synthesis — but board figures have to be bulletproof
Your path through the 4 Levels
Everyone follows the same spine. Here's what each level means at the leadership altitude.
L1
Better Conversations
Read the AUP, learn to prompt for reasoning instead of answers.
For you: build the "trace every figure" reflex before any AI number reaches a board deck or an investor's inbox.L2
Your Workspace
Set up your Claude Teams project and "interview your AI" so it knows your context.
For you: load Company State, the portfolio's KPIs, and your reporting cadence so Claude reasons in our metrics, not generic ones.L3
Run the Workflows
Use the sanctioned Workflow Skills — the reporting cycle, on rails.
For you: the F1 KPI Cycle output and Sam's reporting set are your Level 3 toolbox — the monthly portfolio read, assembled for you.L4
Build New Workflows
Spot the next repetitive task and co-build a new skill for it.
For you: a board-narrative skill, an investor-update skill — the documents you assemble by hand every quarter are Level 4 candidates.Your Skills
Yours = a workflow you own. Source = a Finance-run workflow whose output you consume. Expert = knowledge you summon by topic. All in the Skills Catalog.
Sam's standing monthly reporting set — KPI summary + delinquency view, scope widening as bandwidth allows.
The monthly portfolio KPI read with property-level variance commentary. Mike runs it; you read it — and can ask it follow-ups.
Why a metric moved — variance explanation, benchmarking, same-store trends. Turn a dashboard into a story you can defend.
Score a deal, draft an IC memo, pressure-test underwriting, model value-add and loss-to-lease recapture.
Stress-test the portfolio or a deal against rate, unemployment, and affordability scenarios.
Distribution waterfalls, preferred return, K-1s, NAV, carried interest — the language of LP reporting.
L10 prep, Rock tracking, scorecard design, IDS, accountability chart — run a tighter leadership meeting.
Prompts to try in your voice
Start here, then make them yours. These run in your Claude Teams project with portfolio data loaded.
"Here's this month's KPI dashboard. Give me the 5 things I'd flag to the board — biggest movers, what's improving, what's deteriorating — and mark where you're inferring vs. reading the data directly."
A board-ready shortlist with the AI's guesses labeled, so you know what to double-check.
"Stress-test this acquisition before IC. If the lot-rent ramp is 30% slower and capex runs 20% over, what happens to the return? Show me the downside case, not the pitch."
Forces the bear case onto the table before the room talks itself into the deal.
"Turn this DD packet into a first-draft IC memo using our structure. Leave every figure as a placeholder I have to confirm, and list the open diligence items separately."
A scaffolded memo you fill with verified numbers — never AI-supplied ones.
"Draft the narrative section of the quarterly investor update from these portfolio numbers. Professional, honest about what underperformed, no spin."
A first draft of the hard paragraph — that you then make true.
What NOT to do (Exec / AM edition)
- Board and investor materials are the highest-reputational-risk output we have. Every figure gets traced to source and carries a confidence rating before it goes in a deck. AI assembles; you certify.
- Pre-disclosure data stays out of open chats. Current-quarter actuals, fund-specific capital, and the named pipeline go into a sanctioned Claude Teams project with data you pasted — never "from memory," never a personal account.
- Don't ship the AI's investor narrative unread. It will sound polished and confident even when it's wrong or spinning. The honest version is your job, not the model's.
- Resident-facing strategy still hits Fair Housing. A portfolio-wide rent-increase or screening strategy can't segment by protected class or proxy (income source, family composition, neighborhood). If a plan drifts there, stop and flag it to Carlos.
- No applicant-level screening, even from the top. Don't ask AI to score, approve, deny, or analyze individual applicants or their background/credit — that's prohibited (AUP §2.2). Reviewing aggregate approval rates is fine; tightening criteria based on who they screen out is where it turns into disparate impact — bring counsel/Carlos in.
Your first 30 days
A realistic ramp. You don't have to do it all in week one.
Week 1
Read the AUP and do Personal AI Setup. 30 minutes total. Then use Claude to synthesize five real things — a dashboard, a deal summary, a market question.
Week 2
Set up your Claude Teams workspace. Load Company State and the portfolio KPIs so Claude reasons in our numbers and our cadence.
Week 3
Shadow the reporting workflows. Watch F1 KPI Cycle and Sam's reporting run end-to-end. Ask how each number was derived — that's the muscle that protects the board deck.
Week 4
Run a synthesis solo. Pull the board flags from this month's KPI read, then verify each one against source before you'd put it in front of anyone.
Note for review: The prompt examples and 30-day ramp on this page are a first draft authored from how we already run reporting and IC. Sam and the AM team should gut-check the specifics during the Cohort 2 walkthrough before this is treated as final.