06 execercise

Idea folders existfolder

an AV / adrian-vale folder exists with 2-3 idea subfolders

Usable data filesreal-data

every idea has >=1 data file in a realistic format that parses

Prompt for each ideaprompt-per-idea

every idea has a non-trivial prompt file

Q1Fits the person's role

Judge's reasoning

All three sit in the strategy/corp-dev remit — competitive intel bot and the M&A/market-activation dashboard are Vale's own survey asks, and the AI Transformation Command Center is grounded in the transcript where he and Sulaiman Rao 'are on the strategic initiative teams at Aurex... helping kind of drive AI transformation across the company.'

▸Rubric

Role fit. Is every idea something this specific exec — SVP Strategic Initiatives — would own, rather than another function's job (a marketing campaign calendar, an HR onboarding tracker, an engineering ticket triager) or generic executive-productivity filler (inbox summarizer, meeting-notes cleaner) that any exec at any company could have been handed? Mike: "what I'm looking for here is like, does it come up with interesting ideas that are relevant." FAIL if any one idea sits outside the strategy / corp-dev / cross-company-programs remit.

Q2Specific to the company

Judge's reasoning

Deliverables are saturated with Aurex specifics — 'Clearnet (11,000+ institutions, ~$5T/day messaging network)', 'Parity PUSD circulation tops $2.1B... PUSD growth outpaces AUSD $1.5B', 'Northwind/Aurex Prime $1.25B, Treasura $1B, Tideway $200M, Vaultline', and the messaging-only→settlement conversion motion; none of it survives a company-name swap.

▸Rubric

Company-specific. Are the ideas built on Aurex's actual situation — its products, the AUSD token, the acquisition-integration program, the payments and digital-asset competitive set? FAIL if the deliverable would read identically with the company name swapped for any other mid-size B2B company.

Q3Different ideas

Judge's reasoning

Different problems: outward-facing daily competitor digest + battlecards + reusable bot system prompt (idea 1), internal M&A synergy/partner-funnel/market-license RAG program view (idea 2), and AI adoption + exec follow-through + wins-to-playbook with a CEO update draft (idea 3) — not one artifact re-pointed at new data.

▸Rubric

Ideas are distinct. Do the 2-3 ideas solve different problems for him? FAIL if two of them are the same artifact with different input data (e.g. two dashboards that differ only in subject).

Q4Self-contained prompts

Judge's reasoning

Each PROMPT.md is a complete copy-paste block stating role, competitors/acquisitions, attached files, and 5 numbered deliverables with formats ('produce a <300-word "AV Morning Brief — Mar 10, 2026"'); grep found no `[INSERT`/`<your`/`as we discussed` placeholders anywhere in AV/.

▸Rubric

Prompt is self-contained. Could Vale paste each prompt into a fresh Claude session, with only the files in that idea's folder, and get the thing built without adding anything? FAIL if any prompt has unfilled placeholders (`[INSERT ...]`, `<your company>`), leans on conversation context ("as we discussed", "the ideas above"), or never states what should be produced.

Q5Prompts match the data

Judge's reasoning

Exact two-way match per folder — the eight filenames cited across the prompts (`competitor_signals.csv`, `ma_integration_tracker.csv`, `ai_adoption_pulse.csv`, etc.) are precisely the eight CSVs shipped, each named in its own idea's prompt with a specific job.

▸Rubric

Prompt and data match. Does each prompt name the data files that are actually present in that idea's folder, and does each supplied data file get used by the prompt? FAIL if a prompt references a file that does not exist, or the folder ships data the prompt never mentions.

Q6Realistic practice data

Judge's reasoning

Content is substantive and varied, not props — e.g. 'P-07,Brazil Atlantico,LatAm,Pilot,2025-12-01,22.0,100,Stalled 100d; legal redlines' and dated sourced signals like the 2026-03-03 GENIUS Act reserve-audit guidance; volumes are adequate (42 signals, 28 workstreams, 30 partners, 200 adoption rows) with zero duplicate rows and realistic distributions (13 messaging-only / 10 pilot / 7 settling; 14 Green / 11 Yellow / 3 Red).

▸Rubric

The data is real data, not props. Mike: "the thing I look for here is, um, does it, does it create real data?" Format is already checked by script — judge the *content*. FAIL if values are obvious filler (`Company A`, `Competitor 1`, lorem text, the same row repeated, all-identical dates/amounts) or if there is too little of it to build the thing the prompt asks for (e.g. a trend dashboard shipped with four rows).

Q7Buildable during the session

Judge's reasoning

Every build runs off the bundled CSVs with 'No terminal, no code' constraints and ~35-40 min scopes; live connectors are explicitly deferred — 'After the session: replace CSVs with live connector queries'.

▸Rubric

Buildable in the session. Could this be built during a workshop session against the supplied synthetic data? FAIL if any idea's core function requires live credentials or system access the exec won't have in the room (a real NetSuite/Snowflake connection, his production inbox, a paid market-data feed) instead of working off the included files.