AL/Anthony LaVear

Systems thinker. Hands-on builder.

Good systems.
Better experiences.

I’m Anthony. I bring business systems leadership, AI, and product craft together to make complex ideas easier to understand and use.

Lead Business Systems Analyst & Mentor · Spectrum

CLARIFY THE NEEDSHAPE THE SYSTEMBUILD THE EXPERIENCE

Made & making

Products, tools, web experiences, and explorations.

MOBILE · WELLBEING

LunaCycle

Personal tracking. Local control.
Thoughtful support.

Release preparation

9:41•••

LunaCycle

Your rhythm,
your space.

Sample cycleDay 12

A moment for yourselfLog how you feel

OverviewHistoryLunaLearn

Interface concept · Sample data

Product website ↗

PROJECT NOTES

LunaCycle

Problem solvedCycle dates alone do not explain how someone feels or help a supporter prepare. LunaCycle brings cycle, symptom and emotional-wellbeing context together without requiring a cloud account for core tracking, so personal health information stays under local control.

What’s builtA local-first iOS and Android application with encrypted tracking, profile-aware support and education.

Cycle calendar, flow, symptom and mood logging, forecasts and evidence-based pattern views.
Care Circle and supporter views, practical preparedness and customizable supply checklists.
LunaLearn education, reminders, optional athletic/recovery context and permission-based HealthKit / Health Connect data.
User-selected PDF reports for healthcare conversations, with date-range, included-section and anonymization controls.

What it demonstratesPrivacy architecture, health-data boundaries, mobile UX and practical support workflows carried into a substantial native application.

Current boundaryRelease preparation. Signing, store assets and submission remain open; no public store launch is established. Pattern associations are not diagnoses or proof of causation.

React NativeSQLCipherHealthKitHealth Connect

Product website ↗

WEB · BUSINESS OPERATIONS

NouveauSuite

A connected workspace for providers, booking, calendars, and client care.

Implementation · Release hardening

MON12

TUE13

WED14

THU15

FRI16

Interface concept

Companion app website ↗

PROJECT NOTES

NouveauSuite

Problem solvedIndependent providers need to run appointments, client records, policies and checkout without stitching together disconnected workflows. Built around a small studio’s working day, NouveauSuite gives providers an operational workspace and clients a separate self-service path.

What’s builtA multi-tenant practice platform with provider, client-portal and public-booking applications.

Today and calendar workflows, availability-based booking, rescheduling, client profiles and appointment actions.
Service and policy configuration, forms, checkout adjustments, financial review and Daily Close exception handling.
Tenant-scoped authorization, PostgreSQL row-level security, narrow write capabilities and auditable appointment transitions.
A newer development branch adds dedicated SwiftUI Today, Calendar, Clients, booking/rescheduling, checkout and Daily Close screens—not a web-view wrapper.

What it demonstratesBusiness analysis translated into domain contracts, secure full-stack workflows and a native companion that reuses server-owned rules.

Current boundaryCore P0 implementation is complete; production hardening continues. Native companion and newer management work remain development-branch features. Live native authentication, ordinary-workday/device acceptance, production recovery and provider integrations still need validation; LUXVUE image exchange is not implemented.

ReactNestJSPrismaPostgreSQL · RLS

Companion app website ↗

MOBILE · BEVERAGE DISCOVERY

Drinkfolio

Scan what’s available.
Learn what fits your taste.

Working prototype

Discovery, with a reason.

Barcode capture · Private tasting notes
Explainable recommendations

Product website ↗

PROJECT NOTES

Drinkfolio

Problem solvedChoosing from a shelf or menu is difficult when labels tell you little about your own preferences. Drinkfolio connects what is actually available with private taste history and explainable recommendations, while keeping uncertain identity, source rights and limited evidence visible.

What’s builtA mobile/API discovery loop from product capture and human confirmation to private experience logs, palate signals and comparison.

Barcode and manual entry, canonical product lookup and a review queue for unknown-product contributions.
Private reactions and choose-again notes, category-specific palate inference, confidence labels and deterministic Safe Bet / Best Value / Something New ranking.
Newer development branches connect on-device menu OCR and reviewed matching to Compare, plus published venue browsing, prices and explicit selection.
The latest development slice adds editable/deletable history, firsthand venue reports, operator-reviewed publication, usefulness feedback and account export/closure safeguards.

What it demonstratesSource-aware data modeling, explainable personalization, privacy controls and reliable capture-to-comparison handoffs—not opaque recommendation claims.

Current boundaryWorking prototype with newer beta-development features on separate branches. Production session authentication and combined native/physical-camera acceptance remain open. Venue information is a dated publication, not real-time availability; value comparisons use observed prices, not market-price claims.

Expo · React NativeFastifyPostgreSQLNative OCR

Product website ↗

BEAUTY · VISUAL CONSULTATIONS

LUXVUE

See beauty differently. Preview, customize and save a consultation look—with consent built in.

Active beta development

LUXVUECONSULTATION

Customize this lookClient’s left · Natural · C curl

8inner9region 210region 311region 412region 510outer

Lashes · Brows · LipsPreview look ↗

Interface concept · Sample mapping

Product website ↗

PROJECT NOTES

LUXVUE

Problem solvedA beauty consultation needs more than a verbal description or a lash-map sketch. LUXVUE helps a professional and client explore an estimated look, adjust the mapping and retain consultation context—with explicit permission before storing, capturing or sharing media.

What’s builtAn active beauty-consultation platform with a React professional workspace, API/database foundation and native iPhone implementation.

Editable lash, brow and lip layers: per-eye six-region lash lengths, curl, density, thickness and direction; brow shape, position and color; manual alignment and layer order.
Local photo preview and AR Studio, saved consultation looks, client records and separately recorded storage, capture, export and optional-processing consent.
Five explainable, editable fallback style suggestions; a filter library and Creator Studio with validated, signed free-package import boundaries.
Protected media history, encrypted local recovery, queued consultation notes, reviewed contact import and account/session revocation controls.

What it demonstratesA professional workflow spanning visual editing, native rendering, consent, versioned records and secure offline/media boundaries. The active development branches—not the legacy starter on master—are the relevant product evidence.

Current boundaryActive beta development; native camera/AR hardware acceptance and visual realism are still pending. The suggestions are deterministic, not a trained AI model. Commercial release, marketplace/payouts and delegated NouveauSuite linking remain unfinished. Previews are estimates, not guaranteed cosmetic results.

React · TypeScriptNestJS · PrismaPostgreSQL · RLSCapacitor · Swift · ARKit

Product website ↗

NATIVE IOS · CREATIVE WORKFLOWS

ExactYou

From references and creative directions to image creation and a local gallery.

Development preview

References

Directions

Gallery

Workflow concept

Product website ↗

PROJECT NOTES

ExactYou

Problem solvedPrompt-directed portraits can repeatedly drift away from the person in the references. ExactYou organizes reusable, user-approved likeness photographs and separates identity references from scene direction, so creating another image does not mean rebuilding the same reference workflow.

What’s builtA native iOS creative workspace organized around Studio, My likeness, Gallery and Settings.

Still-photo references, guided five-view enrollment and a separate calibrated dense-capture path; reference ordering, primary selection and notes.
On-device reference-quality guidance, a preview of photos to be sent, categorized directions, saved/recent presets and optional scene references.
Selectable OpenAI / Cloudflare provider paths, output settings and per-request consent that resets when references, notes, scene or provider change.
Protected local capture history and generated gallery, provider filters, reference/result comparison and prompt reuse; credentials stored in Keychain.

What it demonstratesNative camera and creative-product design, provider integration, backward-compatible local persistence and explicit control over what leaves the device.

Current boundaryDevelopment preview. Physical camera/scan acceptance and user-reported generated-likeness quality remain unresolved. Native generation uses photographs, not raw face geometry; no identity-fidelity guarantee or public release is claimed.

SwiftUIAVFoundationARKitKeychain

Product website ↗

AI · REQUIREMENTS GOVERNANCE

GAL Method

Making AI-assisted requirements more explicit, traceable, and trustworthy.

Active alpha

GuideEvidence

AlignDecisions

LeadReadiness

SupportedDerivedUnknown

PROJECT NOTES

GAL Method

Problem solvedAI-assisted requirements can turn assumptions into obligations, lose unresolved questions and confuse a confident answer with verified readiness. GAL makes evidence, human decisions and open questions explicit, with executable controls that keep conversational work aligned to persisted state.

What’s builtGuide · Align · Lead: a requirements method and hardened alpha PowerShell runtime with a host-independent core.

Adaptive intake, evidence classification, provenance and derived-fact discipline.
Persistent questions and phase-sensitive decision debt, with human-owned decisions and explicit readiness boundaries.
Canonical JSON state, schema/version validation, deterministic generated views and explicit migration with backup/rollback safeguards.
Host-adapter contracts, Codex and Kiro adapter sources, runtime/test separation and an allowlisted public-package builder.

What it demonstratesBusiness-analysis governance expressed as repeatable software controls, rather than relying on an assistant’s memory or apparent confidence.

Current boundaryActive alpha. Codex has documented conformance evidence; adapter source availability does not establish tested multi-host parity. Additional host work and release gates remain open. Public source is all-rights-reserved, not open source.

PowerShellJSON SchemaAI governance

Explore GAL on GitHub ↗

More of the work

Web craft, product delivery, and validation.

LUSHAesthetics
Lash & Hair

LUSH Aesthetics Lash & Hair

Implemented site

A responsive service experience with centralized booking and a filterable portfolio.

PROJECT NOTES

LUSH Aesthetics Lash & Hair

Problem solvedA small service business needs a clear path from discovering a treatment to seeing the work and choosing the right booking route. LUSH’s custom site replaces a generic site-builder structure with a focused, brand-consistent service and inquiry experience.

What’s builtA responsive, dependency-light service website in LUSH’s dusty rose and deep brown brand.

Home, About, Hair, Lash, Spa, Gallery, Contact and Thank You pages with shared navigation.
Central booking and social configuration, with service-specific destinations instead of scattered hard-coded links.
Filterable work gallery and lightbox with previous/next and keyboard controls.
Hair-quote form markup, field validation, honeypot and native Netlify POST/thank-you flow; static hosting and asset/security headers.

What it demonstratesService information architecture, responsive front-end craft and maintainable business-site configuration without an application framework.

Current boundaryImplemented site. Current deployment, real form detection/notification delivery and booking destinations were not verified in this review. Gallery imagery must remain approved for public use.

HTMLCSSJavaScriptNetlify

GAL / CHECKConformance

GAL · Codex Validation

Validation project

An isolated runtime and blind-test evidence supporting the GAL Method’s Codex behavior.

PROJECT NOTES

GAL · Codex Validation

Problem solvedA governance method is not credible simply because an assistant says it followed the rules. This supporting validation work isolates the GAL runtime from test answers so host behavior can be evaluated without contaminating the exercise.

What’s builtAn isolated GAL v0.5.0 runtime baseline and documented Codex conformance work—not a separate marketed product.

Runtime/Core, host-adapter contract, state schema, templates and canonical-state/generated-view rules.
Blind-runtime separation: regression fixtures, expected answers and grading records stay outside the evaluation environment.
Controls for evidence provenance, persisted questions, synchronization and validator-before-success claims.
The public evaluation matrix records seven Codex host controls plus one workflow baseline as passing.

What it demonstratesVerification design, reproducible baselines and honest separation of recorded results from execution in the current session.

Current boundarySupporting validation evidence. The historical tests were not rerun for this portfolio review, and the committed runtime alone does not establish a new passing run or other-host parity.

PowerShellHost controlsBlind workflow tests

View the public repository ↗

AL/Systems & craft

anthonylavear.com

Published portfolio

A personal site connecting professional experience with a growing collection of products and technical work.

PROJECT NOTES

anthonylavear.com

Problem solvedA résumé lists roles but does not show how business-systems leadership connects to hands-on product work. This site brings experience, project purpose, implementation depth and readiness together so employers and collaborators can evaluate both the thinking and the craft.

What’s builtA published static Product Studio portfolio with a standalone review preview and preserved résumé and LunaCycle routes.

Nine personal/product showcases and three clearly attributed enterprise tools, with project-specific visual treatments.
Responsive flip-card overviews and notes, keyboard return/Escape behavior, inactive-face focus protection and reduced-motion styling.
Experience and contact controls, existing contact-form integration, and LunaCycle marketing, support and legal pages.
A no-build Git-to-Netlify publishing workflow, preview noindex metadata, production canonical metadata and regression contract tests.

What it demonstratesEvidence-based storytelling, accessible interaction design and maintainable static delivery that connects leadership with working software.

Current boundaryPublished portfolio. Interface concepts are illustrative, not product screenshots. This review does not establish new product launches, employer outcomes or live contact-form delivery.

HTMLCSSJavaScriptNetlify

Enterprise work

Business analysis, built into usable tools.

Tools I designed and built at Spectrum (Charter Communications). Company-owned internal work, presented with interface concepts and sample data. Capabilities and codebase figures are supplied by the author; live systems and internal deliverables are not exposed here.

AI · REQUIREMENTS ENGINEERING

AnalystForge

Raw inputs in. Validated, traceable specifications out.

Active internal platform · v3.3

Spectrum · Charter Communications

AnalystForge ·Work Item Workbench Pipeline ready 94

Pipeline

Upload

Structure

Validate

Deliver

Workbench

Board

Coverage

BACKLOG

US-104 Export to Confluence

INVEST ✓

IN PROGRESS

US-098 Validate NFRs

AC: needs GWT

DONE

US-091 Structure spec

Provenance ✓

Interface concept · sample data

PROJECT NOTES

AnalystForge

What’s builtA self-hosted platform that structures transcripts, documents, decks, and Jira exports into requirements specifications, INVEST-based user stories, BPMN 2.0 process models, and Software Design Document packages. A bounded repair loop applies checks based on five BABOK v3 frameworks. The platform brings together a Kanban workbench with sprints, epics, and bidirectional Jira sync; a searchable, versioned knowledge base; and test traceability. A separate orchestrator assembles specification, test specification, research, implementation plan, and task artifacts in dependency order. Thirty-one JSON tool declarations define its capabilities.

What it demonstratesRequirements discipline expressed as running controls. Eight enforcement checks include six fail-closed checks and two human gates; artifacts require review before publication and carry SHA-256 provenance with JSONL audit entries. Completeness, traceability, confidence, acceptance-criteria validation, version-delta detection, and provenance algorithms run without AI calls. Judgment steps use a configurable local or cloud AI provider. Patent applications filed for several deterministic methods. Live usage and outcomes are internal and not independently shown here.

Author-reported codebaseApproximately 67,500 lines of Python across 182 modules; 24 vanilla-JS front-end modules; 70 unit, property-based, and end-to-end test files.

PythonFlaskSQLite + FTS5Vanilla JSJinja2Git-backed storagepytest / HypothesisMulti-provider AI

AI · BUSINESS ANALYSIS TOOLING

Requirements Specification Generator

A guided pipeline from raw notes to a publication-ready spec.

Internal tool · in use

Spectrum · Charter Communications

Requirements Specification Generator

Transform raw requirements into validated specifications

① Structure

② Validate

③ Assemble

④ Export

BABOK v3 Validation · 5 frameworks

Requirements Validation✓ pass

INVEST sizing⚠ 2 oversized

Traceability matrix✓ complete

Interface concept · sample data

PROJECT NOTES

Requirements Specification Generator

What’s builtA specification generator and companion validation service turn raw notes, Confluence exports, decks, and transcripts into a sixteen-section specification. The six-step workflow covers Structure, Validate, Assemble, Export, Pipeline YAML, and Post-generation check. It extracts use cases, business rules, functional and nonfunctional requirements, and dependencies; applies five BABOK v3 framework checks; and assembles traceability, test scenarios, and sign-off sections. Exports include Markdown, Confluence Wiki Markup, and machine-readable YAML. Companion APIs support INVEST story-sizing, Mermaid process flows, and specification-versus-Jira gap analysis.

What it demonstratesA rigorous method made usable through stages with visible results. Deterministic YAML schema validation checks the machine-readable handoff without a model call, while a configurable AI backend supports local or cloud processing. The live instance, generated outputs, and usage outcomes are internal and not independently shown here.

Author-reported codebaseApproximately 6,000 lines of Python across 22 modules.

PythonFlaskHTML / CSS / JavaScriptMermaid.jsMulti-provider AIYAML schema validation

AI · PROCESS DOCUMENTATION

Jira Process Documentation Generator

Transcripts and surveys in. Documented processes, cross-checked.

Internal tool · tooling-strategy initiative

Spectrum · Charter Communications

Jira Process Documentation Generator Provider: local

INPUTS

📄 kickoff-transcript.vtt

📊 jira-usage-survey.txt

Generate inventory

VALIDATION REPORT

Processes documented12

Matched to source11 ✓

Needs review1 ⚠

Interface concept · sample data

PROJECT NOTES

Jira Process Documentation Generator

What’s builtA web and command-line tool generates Jira process-inventory documentation and cross-validation reports from meeting transcripts and usage surveys. It parses the inputs, produces structured process documentation, and checks the generated documentation back against the source transcript. Confluence-ready Markdown supports downstream use in Confluence, Word, and SharePoint. Provider options include manual copy-and-paste workflows without API keys and local Ollama for automation. The tool supported an internal Atlassian tooling-strategy initiative that produced executive summaries and decision briefs.

What it demonstratesSource-aware documentation rather than an unchecked first draft. Transcript and survey parsers, file-decoding fallbacks, and configurable AI providers address practical enterprise constraints, including mixed inputs and teammates without direct model access. The initiative’s deliverables, findings, and outcomes remain internal and are not independently shown here.

Author-reported codebaseApproximately 4,900 lines of Python across 24 modules, with a Flask interface and Click-based CLI.

PythonFlaskClick CLITranscript / survey parsersMulti-provider AIConfluence / Word / SharePoint output

The person behind the projects

Clear thinking.
Hands-on execution.

My work at Spectrum leads business systems and mentors analysts. My product work puts those same habits of clarity, care, and delivery discipline into apps, platforms, and digital experiences.

Make sense of the problemBusiness analysis

Shape a useful experienceProduct thinking

Build with clear boundariesSystems & AI

Help people move forwardMentorship

Good systems.
Better experiences.

Made & making

LunaCycle

Your rhythm,
your space.

LunaCycle

NouveauSuite

NouveauSuite

Drinkfolio

Drinkfolio

LUXVUE

LUXVUE

ExactYou

ExactYou

GAL Method

GAL Method

More of the work

LUSH Aesthetics Lash & Hair

LUSH Aesthetics Lash & Hair

GAL · Codex Validation

GAL · Codex Validation

anthonylavear.com

anthonylavear.com

Enterprise work

AnalystForge

AnalystForge

Requirements Specification Generator

Requirements Specification Generator

Jira Process Documentation Generator

Jira Process Documentation Generator

Clear thinking.
Hands-on execution.

Experience & education

Business systems leadership

Learning & recognition

Let’s connect

Good systems.Better experiences.

Made & making

LunaCycle

Your rhythm,your space.

NouveauSuite

Drinkfolio

LUXVUE

ExactYou

GAL Method

More of the work

LUSH Aesthetics Lash & Hair

GAL · Codex Validation

anthonylavear.com

Enterprise work

AnalystForge

Requirements Specification Generator

Jira Process Documentation Generator

Clear thinking.Hands-on execution.

Experience & education

Business systems leadership

Learning & recognition

Let’s connect

Good systems.
Better experiences.

Your rhythm,
your space.

Clear thinking.
Hands-on execution.