Dilemma Board — Cited-Evidence Verdict
Can't decide between two options? Get an evidence-checked verdict on a career, build-vs-buy or strategy dilemma instead of confident opinion. Settles a real dilemma (career move, build-vs-buy, quit-vs-stay, which offer to take) in three phases. Phase 1: three blind lenses (Evidence Skeptic, Steelman Arbiter, Precedent Realist) build and stress-test the case for every option. Phase 2: each lens cross-examines the other two. Phase 3: a board verdict by option, tagging evidence CONFIRMED, CORROBORATED or DISPUTED, ending in one named winner or an honest 'too close to call'. Claims must be checkable; unverifiable ones are labeled unverified. It does not just validate the option you prefer, and it is not professional legal, medical or financial advice.
Example input
Use the Dilemma Board on this (a sample I wrote). I run a 4-person SaaS startup (invoice reminders for freelancers, about 1,800 paying accounts, ~$22k MRR). We're on a managed cloud platform and our monthly bill just hit $3,900, up from $1,700 eight months ago. I'm torn between two options: Option A: stay on the managed cloud and spend two weeks optimizing (reserved instances, cheaper database tier, dropping unused services). My cofounder says this "should cut 40%". Option B: move to two rented bare-metal servers at a flat ~$600/month and run Postgres, queues and deploys ourselves. I believe this "is way cheaper and every serious company eventually does it". Constraints: nobody on the team has run production databases on bare metal. We have no on-call rotation, we promised customers 99.9% uptime in our terms, and we have 11 months of runway. Our traffic is steady, with a small spike at month-end. Please give me a verdict, not a list of pros and cons. Contact: founder@example.com
Example output
## Phase 1 — Independent findings
### 🔍 Evidence Skeptic
- Option A claim, "optimizing should cut 40%" (cofounder): plausible-but-unverified. No audit of the bill is cited. The bill rose from $1,700 to $3,900 in eight months while traffic is described as steady, which suggests waste or config drift, and that is exactly what optimization targets. But no line-item breakdown was provided.
- Option B claim, "~$600/month flat": plausible-but-unverified as a total cost. It covers rent only. It leaves out backups, off-site storage, monitoring, a failover or standby box, and engineer time.
- Option B claim, "every serious company eventually does it": unfounded. Many serious companies stay on managed cloud, so this is a generalization with no basis.
- Fact given by user, verified as context: no one on the team has run production Postgres on bare metal; 99.9% uptime is promised in the terms; 11 months of runway; steady traffic with a month-end spike.
- Strip out the unfounded claims: A keeps "a cost cut of unknown size for two weeks of work", B keeps "a large rent saving with unpriced operational risk".
### ⚖️ Steelman Arbiter
- Best case for A: Using only the user's numbers, a 40% cut takes $3,900 to $2,340/month, saving $1,560/month. Over 11 months of runway that is about $17k, bought with two weeks of effort and no change in operational risk. It is reversible, and it protects the 99.9% promise.
- Best case for B: The same arithmetic gives $3,300/month in savings, about $36k over 11 months before extra costs. The workload is steady and small, which suits fixed-capacity servers, and the month-end spike fits within headroom on dedicated hardware. The knowledge gained is reusable.
- The decisive asymmetry: B saves roughly $1,740/month more than A, but exchanges that for unmeasured outage and data-loss risk in a 4-person team with no on-call. One serious incident could cost more than the savings.
### 🌍 Precedent Realist
- Dropbox moved most of its file storage off AWS S3 to its own infrastructure ("Magic Pocket", completed around 2016) and reported large savings in its 2018 IPO filing. Figures are from memory, so unverified. Structural mismatch: Dropbox had thousands of employees, a dedicated infrastructure team and enormous scale.
- 37signals (Basecamp, HEY) publicly left the cloud in 2022-2023 and wrote about the expected savings. Closer in spirit to a small team, but they already employed experienced operations staff and had far larger spend. Exact savings numbers are unverified here.
…Real output from running the skill on a sample problem, shortened. The skill file itself is not published here.