Solutions Engineer II @ Datadog | Specializing in containers, Kubernetes, and cloud observability. My daily work with infrastructure, observability, cloud and containers led me to explore how those same core principles could improve how domain-specific knowledge is captured and managed—and I'm applying it to my Tax Corrector product.
I'm the founder of Tax Corrector, a Django-based property tax analysis platform that helps Texas homeowners challenge unfair assessments and save thousands on property taxes. With an 87% case-found-to-purchase conversion rate, and a 91% customer-success-rate, Tax Corrector transforms complex real estate market analysis into a personalized and custom Property Protest Report that democratizes expertise previously available only to tax attorneys and real estate professionals.
Here's what I've discovered about myself through this journey: I'm that person who can't help but ask "but why does it work that way?" I thrive on architectural thinking that builds bridges from chaos to interconnected systems. I'm naturally drawn to foundational principles that I can apply across domains because understanding the "why" unlocks so much more than memorizing the "how."
I gravitate toward depth over breadth—preferring to gain proficiency in the foundational patterns of cloud architecture (like Kubernetes networking and AWS event-driven services), web development (such as Django ORM optimization and PostgreSQL indexing), and domain-specific algorithms (including Python web crawlers and Property Report optimizations) to build thicker, more resilient foundations.
I'm a firm believer in spending time upfront building infrastructure that scales and compounds—even when it feels like "extra work" initially. What excites me about web development is how seemingly scattered technologies come together to build something useful. Django's MTV pattern, PostgreSQL's advanced indexing, AWS's event-driven services, JavaScript, CSS, and HTMX's progressive enhancement create a holistic system that makes sense. I see the forest and the trees, then I see the connections between them.
Tax Corrector exemplifies this philosophy: the event-driven architecture with EventBridge, Lambda, and Celery creates a resilient system that handles seasonal demand spikes while maintaining reliability for customers making important financial decisions.
When building products, I strike a balance between MVP (1.0 version ready to ship and impress) versus waiting until things are perfect (which never happens). Tax Corrector 1.0 was absolutely ready to ship and impress—even though I had 2.0 on my mind. And customers had no idea I was already dreaming up 2.0 features while they were experiencing an already solid platform that delivered immediate value.
These architectural principles that power Tax Corrector also guide how I build and organize knowledge.
I apply my personal philosophy of Knowledge as Infrastructure (KaI)—pronounced "K-eye"—to focused learning and business intelligence capture for Tax Corrector. Drawing inspiration from Kubernetes and Infrastructure as Code (IaC), KaI applies those same foundations to capturing, organizing and leveraging knowledge.
KaI helps me build competitive advantages within Tax Corrector by methodically securing domain-specific insights about property assessment patterns, county data structure changes, customer pain points, market analysis refinements, and my personal, deep reflections that document every unique challenge and valuable piece of information along the way.
This creates an irreplaceable edge that will eventually power "Tax Agents"—intelligent systems that provide property tax guidance based on context gathered by KaI.
But KaI doesn't only apply to Tax Corrector—it has a far bigger reach—it also shapes how I approach learning and problem-solving.
Which is a perfect match for me because what genuinely energizes me is taking something fragmented and overwhelming—like the complex world of property tax analysis, or full stack web development—and creating elegant systems that organize, connect, and leverage that knowledge so homeowners can navigate property tax challenges and I can build strong solutions.
My approach begins with deep learning. I'm a book-first learner when possible, then I apply those theories to building real applications.
Right now I'm scaling Tax Corrector 2.0 toward a platform that will serve 1,000+ users by April 2026, transforming from reactive report generation to proactive property intelligence.
Every architectural decision, learning investment, and domain insight I capture serves Tax Corrector's mission of making property tax fairness accessible to all homeowners.
That's what keeps me motivated.
Stay curious and keep digging.
-- Josh