By the end of Session 1
You can decompose a technical capability, name an incumbent baseline, grade the evidence and select the fastest useful experiment.
Downloads
Agenda and learning outcomes
| Time | Session | Written output |
|---|---|---|
| 10:00-11:30 | Science-to-Market Translation Under Technical Uncertainty | TEST, HOLD or KILL decision card |
| 11:30-13:00 | Lunch | |
| 13:00-14:30 | Founder-Led Intelligence From Research to Working Demonstrators | BUILD, RESEARCH, PARK or REJECT worksheet |
| 14:30-15:00 | Break | |
| 15:00-16:30 | Disagreement Before Commitment, presented by Denise | Final gate record and collective amendments |
Each 90-minute session follows the accepted 30/25/25/10 format: framing presentation, anonymized case walkthrough, manual small-group activity, then shareback.
By the end of Session 2
You can distinguish market research, meeting intelligence, mockups, demonstrators and production evidence, then write a safe account-to-demo handoff.
By the end of Session 3
You can preserve independent perspectives, assign decision rights and define the evidence required to reopen a gate.
Across the day
You will test a concrete Operator Collective model instead of inventing one from an empty page.
Short glossary
- Evidence grade
- A declared assessment of how directly a claim is supported. A grade is not a probability and does not replace source review.
- Technical distance
- How far a proposed use case sits from the venture's current technical core: DIRECT, ADAPTATION or PIVOT.
- Pivot cost
- The founder-owned estimate of time, talent, data, architecture change and opportunity cost required to pursue a use case.
- Decision gate
- A written state such as TEST, HOLD or KILL, with one owner and a condition for reopening.
- Human review
- The accountable judgment that approves technical meaning, external action, customer contact, publication or final commitment.
- Demonstrator
- A bounded working flow that tests a proposition. It is not by itself production evidence, client value or product-market fit.
- Non-claim
- An explicit statement of what the current evidence does not establish.
Blank worksheets
The HTML versions are keyboard-readable and print cleanly. The PDF versions are formatted for 11 by 17 inch landscape printing.
Session 1: Vertical Triage
Technical assumption, incumbent baseline, distance, pivot cost, fastest experiment and decision gate.
Open accessible HTML worksheetSession 2: Account-to-Demo
Account evidence, screenshot status, supported use case, demonstrator proof target, non-claim and human approval.
Open accessible HTML worksheetSession 3: The Decision Room
Independent scores, disagreements, vetoes, missing evidence, gate owner and reopening condition.
Open accessible HTML worksheetStarting Operator Collective hypothesis
An independent, sponsor-funded collective can place accountable human operators inside pre-seed and seed deep-tech ventures for bounded eight-week assignments. Governed agentic systems expand the operators' research and prototyping capacity. The model should reduce decision time and founder workload without taking equity, ownership of the startup's work or control of the final decision.
Default operating model
- Three to five ventures, with priority given to quantum startups.
- A real decision, bounded data access and scheduled founder gates.
- Human operators for science-to-market translation, account intelligence, prototyping, independent review and gate ownership.
- Agentic systems limited to bounded research, synthesis, code generation, testing and documentation.
- No autonomous outreach, publication, procurement or deployment decision.
- No equity or material startup fee during the pilot. Startup-specific outputs belong to the startup.
Pilot survival criteria
- At least three of five ventures complete a preregistered decision by the deadline.
- Decision time and founder workload improve against baselines frozen at the start.
- The resulting evidence is used in a real buyer, investor or programme discussion.
- At least two independent sponsor categories express willingness to fund another cohort.
- No unauthorized external action or material data-governance incident occurs.
Sanitized case briefs
These briefs disclose enough structure to support the exercises. Company, customer and identifying account details are removed.
Structured translation of a quantum-inspired chemistry engine
Starting point: a venture described broad application potential for a finite-temperature chemistry engine using structured tensor representations and controlled approximation.
Observed work: the engagement decomposed transferable primitives, generated approximately 25 candidate applications, selected five priority wedges and one deferred baseline, and recorded founder-owned gates.
Evidence boundary: one real methodology and completed article exist. Cross-venture effectiveness and real incubator portfolio outcomes remain open.
From public research and meeting preparation to working demonstrators
Starting point: competing ICP hypotheses, incomplete account knowledge and commercial screenshots at different maturity levels.
Observed work: frozen public-web research, company-to-person joins, strategic meeting preparation, persona-panel objections, literature reviews, publication plans and rapid working demos using extensive open-source reuse and guarded cloud deployment.
Evidence boundary: several manuscripts or preprints remain in preparation. Screenshots, synthetic scenarios and cloud demos do not establish production readiness, client savings or product-market fit.
Independent review before commitment
Starting point: a consequential technical decision was reviewed by three independent model families, while a separate document case assigned sections to the readers most able to reject them.
Observed work: the technical review produced 13 unanimous findings and one isolated disagreement. A later closure review found a contract breach despite green executable tests.
Evidence boundary: the M2PI controlled manuscript found that role specialization and compartmentalization did not outperform shared context by default. Alternative assumptions can change that result.
Operator Collective Test
Complete these four questions at the end of every worksheet.
- Which human operator performed the critical work?
- Which tasks could agentic systems accelerate?
- Which decision must remain human?
- What evidence would prove that this intervention created value?
Public references
- NSF I-Corps: experiential customer discovery and commercialization education.
- Activate Fellowship: long-horizon support for scientist founders.
- Duality: quantum-focused business training, facilities and mentorship.
- The Engine Blueprint: foundational programming for tough-tech founders.
- OPERATORS Collective: hands-on operating support, positioned primarily after product validation or early revenue.
- Operator Collective: operator investor community and advisory network.
After the workshop
Sanitized presentation PDFs, completed public case packs, example worksheets, anonymized discussion themes and the amended collective hypothesis will be added after a final disclosure review.
Expression of interest