Problem
Founders often start with broad ideas but no repeatable way to turn an industry choice into a realistic plan, staffing model, income assumptions, or startup sequence.
R&D case study
NAICS Startup Planning System
Planning system that turns broad startup ideas into structured, traceable business planning steps.
At a glance
Context
R&D
Current state
Prototype
Role
Sole architect and full-stack engineer
Core Constraint
Explainability: every generated recommendation must trace back to a rule, data source, or constraint, not a black-box model
Solution / What I Built
Built as an offline-first planning engine backed by the full NAICS hierarchy. The system combines rules-based role generation, income modeling, dependency-ordered startup procedures, and explainability views so users can inspect why each recommendation was produced.
Results
Working planning engine with rules-based role generation, income modeling, and explainability views across NAICS hierarchy
Architecture
The pipeline is shown as explicit stages so the system boundary is inspectable.
Key system pieces
Rules engine converts industry data into launch-plan structure.
Explainability layers make the output inspectable rather than magical.
Offline-first runtime keeps the system usable without external APIs.
Technical Stack
Next.jsPrismaSQLiteZodRules EngineSnapshot Tests
Applied Relevance
Where the pattern matters
- Workflow analysis
- Operational software design
- Prototype planning
- System architecture review
Proof Surfaces
Available artifacts are labeled directly. Missing visuals stay as placeholders until real screenshots are added.
System Walkthrough
Available nowThe current walkthrough is grounded in how the rules engine turns category data into traceable planning output.
- The planning flow shows how an industry choice becomes a generated operating plan.
- Explainability views make each recommendation inspectable rather than magical.
Architecture / Flow
Available nowThe planning engine diagram on this page shows how data, rules, and output generation are connected.
- NAICS hierarchy feeds the planning engine.
- Rules engine drives role generation, income modeling, and procedural ordering.
- Explainability layer exposes why the system made each recommendation.
Operational Surfaces
Available nowThe system has clear operator surfaces even though it is still at the prototype stage.
- Planning outputs can be inspected and exported.
- Rules trace gives users a way to verify recommendations.
- Offline-first runtime removes dependence on external APIs for core behavior.
Artifacts & Evidence
Available nowCurrent evidence is centered on rules, diagrams, and generated outputs.
- Planning engine diagram on this page.
- Rules trace supporting explainability claims.
- Generated plan artifacts available for later inclusion.
Limitations
What this does not claim
- This page describes the current proof available for the project.
- Additional screenshots, logs, or usage artifacts should be added before making stronger claims.
Next Improvements
Reasonable next steps
- Add stronger screenshots or walkthrough artifacts.
- Document validation checks and edge cases more completely.
- Tighten public write-up as the system matures.
Related Case Studies
More portfolio context.
Prototype / Academic ProjectApplied dashboard prototype
WeatherForge
A collaborative University of St. Thomas SEIS 745 project in which David Braun served as lead developer and primary implementer, turning large NOAA weather datasets into county-level risk views and decision-support reporting.
PythonShinyPlotlyParquet
Read case studyPrototype / Academic ProjectLocal RAG prototype
RAGeATM
A small explainable Retrieval-Augmented Generation prototype that retrieves local evidence first, applies a relevance threshold, and refuses unsupported questions when the corpus does not justify an answer.
PythonTF-IDFCosine SimilarityLocal Retrieval
Read case study