Veil
Veil combines calculated birth charts with symbolic readings and continuity between visits. This case study explains how chart facts, authored interpretation, and permitted personal context remain distinct.
Delivery status is not specified in this case study.
Personalized consumer experiences often become repetitive, invasive, or impossible to explain. In an astrology product, the risk is even clearer: exact astronomical calculations, interpretive storytelling, and personal preference signals can be mixed together until the user cannot tell what is fixed, what is creative, or what stayed private.
What makes it useful: Many astrology apps mix calculation, personalization, and generated narrative into one opaque answer. Veil keeps the exact ephemeris separate from interpretation, preserves local preference privacy, and can explain why a particular presentation path was chosen without changing the underlying chart.
Veil explores an atmospheric, private reading experience with continuity between visits. The engineering separates astronomical calculations from symbolic interpretation; it does not establish scientific prediction or unrestricted question coverage.
Browser-local semantic search and bounded models may interpret questions, rank doctrine, plan a reading, and personalize emphasis. Exact ephemerides, chart facts, consent, vault integrity, retained readings, comparison logic, receipts, and rollback remain deterministic.
Calculate the chart and inspect the astronomical positions that form its fixed foundation.
Ask a supported question and read the symbolic interpretation separately from the calculated facts.
Return to a retained reading or compare a later chart, using only the personal context you have permitted.
I built Veil as an atmospheric consumer product with deterministic chart calculations, carefully paced content, restrained interaction, and a privacy-first local personalization direction. The system treats the ephemeris as exact state and the narrative layer as a separate, bounded presentation choice.
I separated deterministic chart calculations from the content that interprets them. The reading pipeline can use permitted personal context while keeping chart identity, retained readings, and change comparisons inspectable.
The OverLift method keeps deterministic ephemeris calculations separate from interpretive content and local personalization. It can show which astronomical state was calculated, which private preferences influenced presentation, and why one content path was selected, while preventing a language model from rewriting the underlying chart or presenting entertainment as scientific prediction.
The method is reusable. Each product supplies its own facts, rules, execution, and validation. These tiers describe the shared architecture; the project scope above records which capabilities are implemented here.
Why determinism matters here: The chart must not change because a language model prefers another story. Determinism preserves the astronomical state while allowing bounded interpretation and private personalization around it.
Define the authoritative result before adding acceleration, semantics, or agents.
Instrument state, latency, quality, and failure so improvement is observable.
Move expensive or high-frequency work behind clear worker and memory boundaries.
Version messages, snapshots, identities, deadlines, and replay inputs.
Let local retrieval and models propose meaning only inside admitted evidence.
Compare several allowed next actions with explicit risk, cost, fit, and uncertainty.
Challenge the production route without giving the challenger authority.
Promote only measured, replayable paths with fallback and operational receipts.
Run advanced evaluators in shadow, compare them with the control, and graduate only repeatable gains.
CATS chooses a qualified execution tier for the evidence, device, latency, permissions, and failure conditions of the task.
The Semantic Bridge maps a goal to explicit entities, evidence, uncertainty, tools, and permissions. Missing or contradictory information remains visible.
Models may retrieve, compare, or propose. Exact software checks facts, permissions, state changes, and execution, then records what happened.
The case study describes calculated charts, retained readings, and comparison over time. Those mechanisms support an inspectable experience; they do not establish that every horoscope question is understood. The portfolio page does not run the Veil application.
