I build practical artificial intelligence tools for fundraising, grant strategy, compliance, digital health, and public-interest decision support.

This portfolio is proof of the Built By User idea: that people close to the problem can turn judgment into working prototypes with AI, then make the work more transparent through public source code, source links, and clear human-review boundaries.

Many of these tools are open for inspection on my GitHub profile. The code is not a claim that every prototype is production-ready. It is a record of how the tools work, what assumptions they make, and where human judgment still belongs.

Federal Acquisition and Compliance

• ClauseFinder
  Searches FAR, DFARS, DAFFARS, eCFR snapshots, proposed rules, and deviation material to help acquisition staff and contractors identify candidate clauses, prescriptions, source history, and applicability signals. It is decision support, not a compliance verdict. Source code.
• Proposal Compliance Checker
  Compares a proposal draft against a solicitation and flags missing, non-compliant, or human-review items. It is built to catch completeness and compliance gaps before submission, not to judge proposal quality or predict funding. Source code.
• NIH Subaward Checker
  Turns a narrow NIH prior-approval notice into a practical grants-office check for domestic subaward changes. It separates deterministic policy logic from AI-assisted extraction and drafting, so a grants officer remains responsible for the final action. Source code.

Fundraising and Grant Intelligence

• Funder Discovery
  Helps nonprofit teams discover possible funders, score fit against an organization profile, review public evidence, and avoid funders that look attractive in a raw search but weak in practice. Source code.
• Funder Finder
  Turns a focus area or proposal draft into a linked set of federal and foundation opportunities, with fit signals instead of a raw search dump. It is designed for the first pass of prospecting, when the real question is where human fundraising time is worth spending. Source code.
• Grant Fit Auditor
  Reviews a grant opportunity against a nonprofit profile, extracts requirements, ranks gaps, and recommends whether to pursue, pause, or decline. The useful argument is refusal: AI should help teams apply to fewer bad-fit grants. Source code.
• Grant Decider
  Gives nonprofit teams a sharper first-pass decision on whether to pursue, pause, or decline a grant opportunity before staff spend scarce time writing. Source code.
• Grant Guardian Analyzer
  Reviews nonprofit financial risk signals so program and fundraising teams can ask better questions before a grant moves forward. It is designed as a due-diligence aid, not a substitute for judgment from finance, program, or leadership teams.
• Funder Discovery Plugin
  Codex plugin and Custom GPT Actions pilot for nonprofit funder discovery, fit scoring, 990-backed funder briefs, donor prospect briefs, and CRM-ready pipeline exports. It is a workflow prototype for fundraisers, not a replacement for relationship strategy. Source code.
• Foundation Grant Signal Lab
  Turns a grant proposal, applicant profile, and foundation strategy into a first-pass decision memo, with the claim, score, review route, board line, visible evidence, funder risks, and next diligence actions. Built for foundation staff who need to separate polished proposal language from defensible grantmaking judgment. Source code.

AI Workflow and Knowledge Tools

• Claude Skills Creator
  Helps teams turn repeatable expertise into structured AI workflows, including grant research, grant compliance review, credibility audits, and funder-specific writing voice. Source code.
• Network Radar
  Local-first relationship search for people who make introductions, shortlists, and referrals from a large personal network. It keeps contact data, Gmail metadata, LinkedIn results, OAuth tokens, and the SQLite database local by design. Source code.
• Kindora for ChatGPT MCP
  Python MCP server that exposes Kindora’s public funder and grant tools as a ChatGPT custom connector. It is a thin, read-only proxy for grant discovery workflows, with tests and one-click hosting configuration. Source code.

Healthcare, Public Affairs, and Strategy

• AI Healthcare Contract Simulator
  Models how the same AI documentation product behaves differently under fee-for-service, MSSP, Medicare Advantage, and employer-direct contracts. The goal is to make incentives visible before a digital health team commits to a business model. Source code.
• Healthcare Testimony Strategy
  Helps healthcare executives and public affairs teams prepare for Senate testimony by aligning claims with committee priorities, likely senator concerns, public evidence, and safer answer frames. Source code.
• Committee Analysis
  Analyzes congressional committee context so testimony, advocacy, and public-interest messaging can be prepared against the actual audience, not a generic policy persona.
• Collins Public Record RAG
  Local citation-first research system for Senator Susan Collins’s public record over a rolling five-year window. It ingests official materials, stores source records, chunks and embeds them, and answers only from retrieved cited evidence.
• Senator Profiling System
  Local evidence-first system for building reusable U.S. senator profiles and cited testimony guidance from public records. It labels observed facts, inferences, and recommendations so analysts can review the evidence before use.

Digital Health and Caregiver Tools

• First Signs Health
  Turns published cognitive screening protocols into a browser-based, family-usable workflow that produces a clinician-ready handoff document. It does not diagnose. It helps a family prepare for a better medical appointment.
• Dementia Care Billing Guide
  Translates dementia-care billing complexity into a more usable decision aid for clinicians, caregivers, and families trying to understand care options and documentation needs.
• Cognitive Screening Resource
  Explores how simple browser-based tools can make cognitive screening workflows more understandable while keeping the medical decision in clinical hands.

AI Search, Education, and Public Tools

• GEO Audit Studio
  Creates a generative engine optimization audit prompt, scorecard, measurement plan, and action pack so website owners can test whether AI tools can access, understand, cite, and accurately describe their site. Source code.
• College Ranking Tool
  Tests how AI can make ranking logic more transparent by letting users see the criteria and tradeoffs behind a recommendation instead of accepting a black-box list.

How I Build

Built by User is my build philosophy. Software no longer requires a six-month procurement cycle before anyone tests a core idea. A domain expert can now prototype the useful 20 percent in days, then bring in engineering judgment where the work touches sensitive data, production systems, or consequential decisions.

I start with a concrete decision that wastes time or money today: which funder to pursue, whether a proposal is worth writing, what a compliance notice actually requires, what clause may apply to a contract, or what evidence a clinician needs. I build the smallest useful tool around that decision, keep links to source material visible, and separate prototype value from production readiness.

The source code for selected projects is available at github.com/wayanvota. Public code is part of the accountability model: it makes the method inspectable, the limits easier to see, and the next version easier to improve.