Project

Career Decision Evidence Engine

An explainable decision interface for graduate study, employment, international paths, entrepreneurship, and parallel strategies.

HTML · JavaScript · LocalStorage · Decision Modeling · O*NET

2026

Product Snapshot

Role: Independent product prototype
Users: People comparing several career paths who need evolving evidence rather than a one-time score.
Stage: Interactive local prototype completed; the public description removes all personal decision data.
Focus: Multi-path comparison, O*NET assessments, weights, evidence networks, sensitivity, and export.
Validation: Implemented prefilled structures, interactive scoring, linked evidence, local persistence, export, and parallel-strategy support.
Public Proof: Working local HTML/JavaScript application; GPA, family, assets, interests, and other private inputs are excluded.

Next Step

Portfolio PDF About the Founder Email

Decision problem

Career choices are often reduced to one binary question. Real decisions combine time, opportunity cost, risk, identity, skill accumulation, and the option to pursue paths in parallel. One total score can create false certainty.

This prototype places graduate study, employment, international paths, entrepreneurship, and parallel strategies in one adjustable model.

Core interaction

Users can update evidence, change criteria and weights, incorporate O*NET-style assessments, inspect supporting and conflicting links, persist state locally, and export the analysis.

Parallel strategies are treated as valid choices rather than evidence that a decision has failed.

Privacy boundary

The original local version contains real personal inputs. This public record describes only the system. It excludes grades, family conditions, assets, relationships, and preference details, and it does not expose an unsanitized demo.

Product judgment

The goal is not to produce an answer. It is to make “why I believe this path” inspectable and revisable. The prototype directly informed the decision-support direction in Life Exchange.

Current limitation

Weights still come from the user and cannot replace interviews, opportunity tests, or longitudinal tracking. The next useful step is measuring prediction-versus-outcome error, not adding more criteria.