This path curates the critical inflection points in evolving a physical (or physical + embedded software) project into something real customers can buy repeatedly. Each stage lists core objective, gate / exit criteria, and AI assist patterns. Use it as a governance rail: do not advance until the gate is objectively met; use AI to reduce iteration cycle time, not to skip validation.
1. Vision & Problem Framing
Objective: Articulate a persistent pain (who / situation / costly friction) and transformation promise measurable in clear metrics (time saved, error reduction, reliability lift).
Gate: One-line problem + quantified success metric + list of failed current workarounds.
AI Assist: Cluster interview transcripts; extract repeated nouns/verbs; generate negative persona (who it is not for) to sharpen focus.
2. Audience & Market Validation
Objective: Evidence of willingness to pay / adopt from early adopter profile; articulate JTBD (Job-To-Be-Done) statements.
Gate: ≥10 convergent interviews + pre-commit signals (LOIs, pilot signup list) + pricing hypothesis range.
AI Assist: Transcribe + summarize calls; objection clustering; JTBD template filler; pricing sensitivity scenario generation with synthetic customer archetypes.
3. Prior Art & Differentiation
Objective: Ensure novelty / defensibility and find whitespace; map features vs incumbents, open hardware, patents.
Gate: Differentiation matrix; no fatal patent landmine; unique value narrative.
AI Assist: Patent abstract summarization; semantic similarity search; competitor claim contrast generation.
4. Requirements & Specification
Objective: Translate vision into testable functional + non‑functional spec (performance, safety, power, latency, environment).
Gate: Ambiguity lint returns low vagueness; invariants enumerated; acceptance tests scaffolded.
AI Assist: Spec linting (weak verbs, fuzzy adjectives); edge case enumerator; risk list generation; interface contract drafting.
5. Rapid Prototyping
Objective: Retire physics or interaction unknowns with minimal spend (looks-like & works-like prototypes).
Gate: Critical feasibility questions answered; updated risk register shrinks top unknowns.
AI Assist: Parametric CAD suggestion; simulation parameter sweeps; BOM auto extraction; alternative material prompts.
6. Component Sourcing & Supply Risk
Objective: Assemble resilient preliminary BOM with lead-time & obsolescence insight.
Gate: ≥2 vetted suppliers for every critical component; risk heat map.
AI Assist: Datasheet summarization; compatible alternative part search; supply risk scoring; MOQ / lead-time normalization.
7. Cost Modeling & Unit Economics
Objective: Validate margin path at target volume tiers.
Gate: COGS model with sensitivity; gross margin within strategic band.
AI Assist: Dynamic cost rollups from BOM; scenario simulation (volume / commodity price shifts); break-even calculator templates.
8. IP Strategy (Patent / Trade Secret)
Objective: Prioritize what to patent vs keep internal; defensive publication list.
Gate: Draft claim scaffold or conscious decision to defer; conflict scan clean.
AI Assist: Claim variant drafting; novelty mapping; infringement proximity summaries.
9. Regulatory & Compliance Scoping
Objective: Identify applicable directives early (FCC / CE / UL / RoHS / REACH / FDA / OSHA / GDPR etc.).
Gate: Standards applicability matrix + required test list + cost/time estimate.
AI Assist: Standards text classifier; extraction of mandatory clauses; delta diff when standards update.
10. Certification Path Planning
Objective: Construct sequencing for pre-compliance tests, formal lab tests, documentation readiness.
Gate: Timeline covering safety, EMC, environmental + owned responsibility matrix.
AI Assist: Auto-generate test procedure outlines; documentation checklist ingestion; gap detection vs standard clauses.
11. Testing Strategy & Telemetry Design
Objective: Comprehensive test matrix (bench, environmental, reliability, usability) + instrumentation schema for logs.
Gate: Coverage map; telemetry schema supports every KPI / failure hypothesis.
AI Assist: Generate test cases from spec; anomaly detection in sensor logs; failure clustering; automated summary of test runs.
12. DFM / DFA
Objective: Reduce part count, assembly time, tolerance stack risk.
Gate: Revised CAD + assembly steps + yield/cycle projections.
AI Assist: Part consolidation suggestions; tolerance analysis; assembly instruction drafting with step images placeholders.
13. Pilot Deployment
Objective: Real-world usage to validate retention, reliability, and value metric delta.
Gate: Pilot KPI thresholds (retention %, failure rate, NPS) met or actionable deltas identified.
AI Assist: Cohort analysis; sentiment extraction; usage clustering; root cause suggestion from defect narratives.
14. Manufacturing Partner & QMS
Objective: Secure scalable production + quality management system (SOPs, NCR, CAPA).
Gate: Vendor scorecard selection; QMS artifacts baseline; revision control process live.
AI Assist: RFP comparison; contract clause risk extraction; NCR categorization; anomaly alerts on production telemetry.
15. Security & Firmware / Software Updates
Objective: Maintain trust for connected devices (secure boot, OTA pipeline, SBOM).
Gate: Threat model updated; patch cadence defined; vulnerability triage SLA.
AI Assist: Static analysis triage; CVE mapping from SBOM; threat scenario generation; patch note summarization.
16. Packaging, Logistics & Sustainability
Objective: Safe delivery + sustainable material choices + clear end-of-life guidance.
Gate: Drop test pass; recycling instructions; carbon estimate documented.
AI Assist: Material alternative suggestions; packaging failure prediction from prior datasets; sustainability claim drafting.
17. Go-To-Market & Pricing
Objective: Positioning, message hierarchy, pricing ladder, channel plan.
Gate: Distinct competitive slot; channel margin model feasible; early adopter narrative resonance tested.
AI Assist: Message variant generation; competitor claim contrast; elasticity scenario modeling.
18. Funding & Capital Stack
Objective: Milestone-based budget fueling risk retirement sequence.
Gate: Runway covers next two major uncertainty reductions; dilution scenarios clear.
AI Assist: Cashflow forecasting; pitch deck refinement; cap table simulation; burn anomaly alerts.
19. Launch Readiness Gate
Objective: Verify supply buffer, support scripts, monitoring, risk register mitigations.
Gate: All high severity risks with mitigation owners & deadlines; support SLA instrumentation live.
AI Assist: Risk clustering; unresolved mitigation summarization; checklist coverage diff; FAQ answer drafting.
20. Post-Launch Support & Feedback Loops
Objective: Sustain satisfaction & accelerate learning from real issues.
Gate: Mean resolution time within target; feedback taxonomy saturating.
AI Assist: Ticket intent classification; auto-draft responses; escalation predictor; churn risk scoring.
21. Roadmap Governance
Objective: Prioritize evolutions tied to metrics or validated insights (not vanity).
Gate: Each backlog item tags: metric, hypothesis, evidence source.
AI Assist: Backlog deduplication; dependency graphing; feature impact estimation.
22. Lifecycle & End-of-Life
Objective: Plan responsible retirement, refresh, and part availability.
Gate: EOL criteria defined; replacement path doc; sustainability commitments mapped.
AI Assist: Replacement part demand forecast; recyclability analysis; upgrade migration instruction drafting.
23. Continuous Improvement Metrics Layer
Metrics: Time-to-prototype, iteration cycle time, defect density, yield %, pilot retention, NPS, margin trajectory, convergence iterations per spec, evaluator coverage %, exploration yield.
AI Assist: Metric anomaly alerts; causal inference suggestion; dashboard narrative summarization; KPI correlation exploration.
Pattern Summary
Loop: Input artifacts → AI augment (summarize / propose / cluster / predict) → Human judgment & pruning → Logged rationale → Update canonical spec / repository. Disciplined loops, not prompt dumps, compound advantage.