What I Do
WCAG Audits & Technical Remediation
I find accessibility issues including the ones Lighthouse, AI, and automated tools miss, and I fix them. Not just a report of findings for your team to interpret. Actual pull requests, component fixes, and tested solutions in your codebase. Your team gets the engineering context with the finding, so they can ship the fix, not just understand it.
Accessibility & Responsible-AI Engineering
I either embed with your team as your accessibility and responsible-AI engineering lead, or I take on the engineering myself for greenfield builds, rebuilds, and full-product engagements. I audit, set priorities, open PRs, review code, ship the work, and build accessibility and responsible-AI patterns into the system from day one. I work across levels and domains until your team is self-driven, or, when I'm the one building, until the product is shipped right. If you already have AI in the product, I'll help you make it more responsible, ethical, and inclusive. If you don't yet, I'll help you build it right from the start. I'll help you pitch it to leadership too. Honestly, it's an easy sell for anyone serious about real business outcomes.
Results
Gap Inc.
I embedded across domains, teams, squads, and brands across Gap and drove accessibility from afterthought into the home-page team's product-engineering workflow.
- Authored and presented an accessibility strategy proposal projected to sustainably generate $300M-600M+ annual net revenue.
- Audited and remediated accessibility issues across the home page with hundreds of millions of visits and tens of millions of users.
- Led accessibility in two multi-million dollar engineering projects: Rapid Authoring CMS and 2.0 CMS components.
- Caught, fixed, and led remediation for accessibility issues on design system and codebase that Lighthouse, axe-tools, designers, engineers, and QA did not flag.
- Built an automated Lighthouse testing suite with AI/LLMs in a single sprint, finding unknown performance improvements and justifying investment in a multi-million dollar project (2.0 CMS components).
- Improved Largest Contentful Paint by 25% and Cumulative Layout Shift by 99% on home page across all 6 brands with over 1 billion+ views and 100M+ users.
- Established a minimum of 1-2 accessibility tickets allocated to every sprint at the home-page team, making accessibility a recurring part of the engineering workflow rather than a periodic afterthought.
Independent Methodology Research
I built a first-principles evaluation framework against 6 frontier model families and surfaced a finding that named-brand AI safety teams had not published.
- Caught self-identity hallucinations in 4 of 6 frontier models: Anthropic, OpenAI, Google, and Mistral.
- This is the same engineering rigor I bring to accessibility audits and remediation in modern stacks, including codebases where AI features are part of the surface I'm reviewing.
- Not a hypothesis. A reproducible finding.
Before focusing on accessibility, I spent 8 years shipping production code across the full stack: microfrontend architectures at Verizon, production GraphQL APIs at Odeko, a pilot and production PWA for 100 B2B customers that helped raise $14M in seed funding at SumAll, and a component library used across three apps. Before I joined official engineering teams, I built the first instant motorcycle service quoting tool and on-demand mobile motorcycle mechanic service on the web for my startup, motofix. I know how engineering teams and tech businesses work because I've been on them for over a decade.
Why Me
I'm a developer, not an auditor.
I sit with the code and the models. I run tests, write findings, ship fixes. The deliverable is working software, not a PDF you need to interpret. 10+ years writing production code across the frontend, backend, and full stack in teams of every size and stage. I treat your business like my own and your accessibility problems like they're humanity's, because they are. When I find an issue, I don't just flag it. I open a PR, fix it, and drive it to resolution across teams and domains if that's what it takes. I own it end to end and build that same ownership into your system and team. My job is only done once your engineering culture is truly inclusive and responsible AI oriented. When something is genuinely outside my expertise, I partner with domain experts in digital accessibility as needed. That's part of the practice, not a workaround.
I have lived neurodivergent insight.
I'm neurodivergent and autistic. I experience the barriers that automated tools, neurotypical testers, and even dedicated accessibility teams miss. Cognitive overload. Sensory friction. Interaction patterns that technically pass WCAG or AI scanning but fail real humans. When I test your product, I'm not simulating the user experience. I'm having it. Frankly, this isn't something I could earn through certification.
I treat AI as a craft, not a marketing layer.
When your stack includes AI features, I review them with the same engineering rigor I bring to anything else. Testing for failure modes. Building patterns that don't break under real conditions. Refusing to ship what doesn't work. I built a 9-model first-principles LLM evaluation framework that caught self-identity hallucinations in 4 of 6 frontier models. Anthropic, OpenAI, Google, Mistral. I take AI behavior seriously, and that seriousness shows up in how I engineer accessibility on AI-touched surfaces.
- Intro to Web Accessibility - W3Cx
- Professional Member, International Association of Accessibility Professionals (IAAP)
- Contributor in W3C Community Group: COGA (Cognitive and Learning Disabilities Accessibility)
- 10+ years professional web engineering experience
- Coursework in Information Systems, CUNY School of Professional Studies
- Coursework in Psychology with concentration in Behavioral Neuroscience, CUNY Hunter College
Interested?
If you're serious about building truly accessible products, AI systems, or teams, I can help. Book a free 30-minute consultation. I'll listen first. If I'm the right person for the work, we'll talk through how it would actually run. If I'm not, I'll tell you and point you toward someone who is. Honestly, that's the practice.
Email me