Soulful Tech
Where Soul Meets Software
Full-stack products, AI integration, and a security posture built in from the first line of code. Twenty years of the work, shaped by the belief that technology should serve the person on the other side of the screen.
Selected work
Three products, one standard.
Each one answers to the same question. Does this actually serve the person using it?
VeloxSync for Education
K-12 AI platform with grade-band intelligence. 10 DB tables, 30+ endpoints, 112 state standards, Ei-Core fine-tuned models.
K-12 · 112 state standards
Meraki Lingua
37-language AI chat widget. Dialect-accurate, RTL-supported, Claude API-powered. HTML PDF export built in.
37 languages · RTL support
Canopy Guard
Client security posture auditor. MITRE ATT&CK mapped findings. CISSP-informed. Branded client-ready reports.
CISSP-informed · ATT&CK mapped
What I do
Four ways to build with intention.
Product Engineering
I design and ship full-stack products end to end, from the Postgres schema through the typed API to the React interface, treating them as one coherent system rather than three handoffs. The stack is Next.js, TypeScript, and Tailwind on the front, with Supabase, Postgres, and Railway carrying the data and deployment. The outcome is software that stays fast and predictable under real production load, with a data model that does not buckle the first time traffic or scope grows.
VeloxSync runs 10 database tables, 30 plus endpoints, and 112 state standards in production.
What you get
- Next.js, React, and TypeScript
- Database design and API architecture
- Supabase, Postgres, and Railway
AI Integration
I bring large language models into products in a way that earns trust, integrating the Claude and OpenAI APIs, adding retrieval-augmented context, and fine-tuning only where the data clearly justifies it. Every integration ships with guardrails, evaluation, and fallbacks so the output is something you can put in front of a client without flinching. The result is AI features that handle real conversation, including multilingual and right-to-left support, instead of demos that fall apart at the edges.
Meraki Lingua holds conversation across 37 language communities, dialect by dialect, on the Claude API.
What you get
- Claude and OpenAI integration
- Fine-tuned and retrieval-augmented systems
- Multilingual and RTL support
Security Posture
Security is part of the architecture from the first commit, not a checklist bolted on the week before launch. I bring a CISSP lens to the threat model, the dependency tree, the headers, the TLS configuration, and the data-handling paths, then map every finding to the MITRE ATT&CK framework so the risk is concrete. You walk away with a client-ready posture report and a hardened system, the same depth of review enterprises routinely pay six figures to commission.
Canopy Guard maps findings to MITRE ATT&CK and delivers the posture report enterprises pay six figures for.
What you get
- CISSP-informed architecture review
- MITRE ATT&CK mapped findings
- Client-ready posture reports
Content & Strategy
Good software still needs the right words and the right direction around it. I write the technical documentation, the product narrative, and the brand voice, then help you aim the product at the outcome that actually matters to the business. Drawing on four published books and twenty years across leadership, IT, and information security, I turn a vague goal into positioning and messaging a team can rally behind and a customer can understand.
Four published books and twenty years across leadership, IT, and information security.
What you get
- Technical writing and documentation
- Product strategy and positioning
- Brand voice and messaging
At a Glance
Two decades, measured.
The numbers behind the studio. Twenty years of work, a handful of products carrying real load in production, and the credentials to stand behind the security claims. Here is the shape of it at a glance.
| Metric | Figure |
|---|---|
| Years of Experience | 20+ |
| Products in Production | 5 |
| Languages Supported | 37 |
| State Standards Seeded | 112 |
| Security Certifications | CISSP |
| Clients Served | 10+ |
How the work gets made
A process you can actually see.
Most engagements move through the same four movements, and the reason they are written down is simple. You should never have to wonder what is happening to your project or why. Every phase ends with something concrete in your hands, whether that is an architecture diagram, a working preview, or a hardened deployment you can put in front of customers. The pace is deliberate, the feedback loops are short, and there is no black box between the first conversation and the thing that ships.
- 01
Discovery
We start in conversation, not in code. I learn the people who will use the thing, the constraints you are actually operating under, and the real goal hiding behind the stated one. Nothing gets proposed until I understand the problem well enough to argue both sides of it.
- 02
Architecture
I map the system before building it. The data model, the integrations, the failure modes, and the security posture are decided up front, on paper, where changing your mind is cheap. You see the shape of the whole product before a single migration is written.
- 03
Build
The work gets made in tight, visible loops. You get previews you can click through, progress you can measure against the plan, and a direct line to the person writing the code. No status theater, no surprises three weeks before launch.
- 04
Deploy and Harden
We ship, and then we secure. Monitoring goes in, headers and TLS get tightened, dependencies get audited, and the system is reviewed through a CISSP lens before it carries real traffic. You launch with a posture you can defend, not a promise to fix it later.
That last movement is where most studios quietly cut corners and where twenty years in information security refuses to let me. A product is not done when it works on a good day. It is done when it holds up on a bad one, when the traffic spikes, when someone probes the endpoints, and when the data it was trusted with stays exactly where it belongs. Building that kind of resilience in from the start costs a little more attention early and saves an enormous amount of pain later, and it is the through line that connects every project this studio has shipped.
The same discipline shows up in the smaller decisions, the ones a client rarely sees but always feels. A form that tells you exactly what went wrong instead of failing silently. An API that returns a clear, structured error instead of a stack trace. A page that loads fast on a tired phone and a slow connection, because the person on the other side of the screen did not ask to be punished for their hardware. These are not features anyone puts on a roadmap, and they are precisely the things that separate software people tolerate from software people trust. I treat them as part of the job, not as polish to add if the budget survives.
Working with a studio of one has a specific shape worth naming. You are not handed to a junior team after the pitch, and the person who understood your problem in the first call is the person writing the code in the last one. That continuity means less is lost in translation, decisions get made faster, and the context you spent an hour explaining does not evaporate between meetings. It also means I take on the work I can genuinely do well and say so plainly when something falls outside that, because a good referral protects your project more than an honest overreach ever could. The studio stays small on purpose, so the care stays real.
It also helps to know what kind of work this studio is built for. The best engagements tend to be the ones where the problem is real and a little messy, where someone has tried the obvious solutions and found them wanting, and where quality matters more than shipping the cheapest possible version by Friday. Early-stage founders proving an idea, established teams who need senior hands on a hard subsystem, and organizations that finally want their security posture taken seriously all tend to find the fit natural. If your situation looks nothing like those, that is useful to learn early too, and I will tell you so rather than bend the work into a shape it was never meant to take.
If any of this sounds like the kind of partner your project has been missing, the next step is an unhurried conversation. No deck, no pressure, no obligation on either side. We talk through what you are building, where it hurts, and whether the way I work is the right fit for the problem in front of you. The worst case is an honest answer and a useful pointer in a better direction. The best case is the start of something built with intention from the very first line of code.
Soulful Tech is not a tagline I picked because it sounded warm. It is the only way I know how to build.
The meditation and the Reiki practice are not a hobby that lives next to the technical work. They are the source of the pace, the attention, and the decision to stop and ask whether a feature actually serves the person using it.