Forward-deployed engineering / Product systems / AI workflows
I’m Josh Fletcher, a hands-on software architect and product builder working across business operations, full-stack engineering, and applied AI. I translate customer needs into technical direction, build the solution, and own delivery through rollout and production troubleshooting.
I work best where the problem is real, the path is unclear, and somebody has to connect strategy to implementation.
That has led me from economics and consultative sales into full-stack development, product ownership, automation, AI systems, and the unglamorous production work that makes all of it reliable.
Where I’m heading
I’m pursuing roles where understanding the customer, designing the system, and shipping the solution are part of the same job.
Multi-client platform
A shared outreach platform for Deaton Engineering and SMOD that connects company research, AI-assisted drafting, human review, and client-specific delivery. I own architecture, integrations, deployment, and production troubleshooting.
B2B SaaS
A dealer incentives and enablement platform connecting rewards, product education, and sales engagement. As U.S. partner, I lead product requirements and translate manufacturer and dealer needs into technical delivery and rollout decisions.
AI product venture
A personalized storytelling product designed to help children explore interests and learning goals. As founder and AI product architect, I own product direction and the technical design across user experience, full-stack systems, and AI workflows.
AI development workflow · In development
An AI-assisted software workflow in development that connects project context, repository evidence, and human approval to controlled software changes.
Earlier interface experiments and learning projects remain available on GitHub. The work above better represents the problems I solve now.
I connect product, engineering, and operations to make decisions across the whole system.
I work with stakeholders and specialists to turn those decisions into maintainable software.
I move from an unclear need to a usable product, keeping business logic, interface design, and implementation connected.
I connect services, data, and workflows into systems people can actually operate and troubleshoot.
I use AI where it creates leverage, then add the review, evidence, and failure boundaries required for production work.
TypeScript · Next.js · React · Node.js · Express · MongoDB · AWS · Docker · GitHub Actions
I clarify the operational problem, define success, and choose a practical path to delivery.
I reduce uncertainty, define what “working” means, and create a path that stays verifiable as the system grows.
“The job is not to make software look finished. The job is to make the operation work better.”
Separate the visible request from the operational problem underneath it.
Trace people, data, decisions, dependencies, and failure paths before adding machinery.
Ship an end-to-end slice that can be used, observed, and improved.
Make the system explain what happened and stop safely when certainty runs out.
Validate in the real environment, learn from friction, and keep the product maintainable.
Have a difficult system to untangle?
I’m open to forward-deployed, full-stack, solutions architecture, and technical product opportunities, along with focused partnerships where I can own the path from problem to production.