Wallboards Leaderboard v2
Rebuilt the real-time Leaderboard slide for a contact-center Wallboards display system in six weeks with one teammate — new UI, better slide management — as part of a 12-engineer platform-wide replacement effort.
Daniel Boyd — Software Engineer
Full-stack engineer building software that holds up in production.
Two-plus years shipping and maintaining features in a large enterprise platform (.NET, Angular) — plus a growing habit of building directly with LLMs, most recently a retrieval-augmented assistant for tabletop game masters.
RoleSoftware Engineer
FocusFull-stack delivery, LLM/RAG
StackTypeScript · C# · Python · Angular
CurrentlyShipping an AI assistant for DMs
Selected work
Projects and product-focused delivery
AI Dungeon Master Assistant
Deploying nowAn LLM-powered assistant for tabletop game masters covering spell and monster lookup, encounter building, and full worldbuilding support, backed by a retrieval-augmented generation pipeline over game rules and lore.
Enterprise Localization
Extended a large enterprise product to Simplified Chinese, Traditional Chinese, and Korean, with groundwork laid for Hebrew ahead of a pending regional deal, plus contributions to a shared internal localization tool.
Experience
Professional engineering impact
Software Engineer · NiCE
- Led a 6-week rebuild of the real-time Leaderboard wallboard slide (v2) with one teammate, delivering an improved UI and slide-management system as part of a 12-engineer, platform-wide Wallboards replacement.
- Resolved 30+ production issues across frontend, backend, and data layers — including accessibility gaps and broken sub-modal behavior — improving stability and reducing customer-reported defects.
- Expanded product localization to Simplified Chinese, Traditional Chinese, and Korean, and laid groundwork for Hebrew ahead of an upcoming regional deal; contributed to an internal localization tool used across engineering.
- Contributed to a large-scale frontend framework refactor, improving code quality, consistency, and long-term maintainability across the codebase.
- Contributed to an early performance-tuning project, during ramp-up on the platform, that helped cut dashboard presentation load time by roughly 80%.
- Use agentic AI coding tools daily to accelerate feature development, debugging, and code review in a production environment.
- Collaborate with product managers and cross-functional partners to scope work, break projects into phased deliverables, and provide feedback through code reviews.
Skills
Core technologies
TypeScriptC#PythonMySQLAngularNext.jsFlaskReactOpenAI APIRAGKotlinFlutterREST APIsGitJiraAgile Delivery
AI proficiency
How I use AI in real software delivery
Building with LLMs directly
Designing and shipping a retrieval-augmented generation assistant for tabletop game masters, using OpenAI models over a custom knowledge base of rules and lore.
AI-augmented implementation
Use Claude and GitHub Copilot daily in production work to accelerate scaffolding, refactoring, and repetitive implementation tasks while preserving code quality.
Review and validation mindset
Treat AI output as a draft: verify behavior, align with architecture, and ship only after measurable confidence through testing and review.
About me
Engineering style and collaboration
I’m strongest when translating broad goals into working software with clear tradeoffs. I care about dependable execution: diagnosing issues quickly, reducing complexity, and keeping delivery predictable.
Outside work, I run tabletop campaigns as a dungeon master. That’s part of why I started building an AI assistant to help other DMs run their own tables — combining a hobby I love with hands-on LLM engineering.
Focus areas
- Full-stack feature delivery with strong debugging discipline
- Performance optimization and UX-focused product improvements
- Localization, maintainability, and team collaboration in Agile workflows
Off the clock
- Class: Dungeon Master
- Current quest: shipping the AI DM assistant
For fun
FFXIV progression with live FFLogs data
I keep this section because passion projects still show engineering habits: integrating third-party APIs, handling auth securely, and presenting noisy data clearly.
Data comes from a server-side Next.js API route using FFLogs OAuth credentials from environment variables. No API secrets are exposed to the browser.
FFLogs
Current Raid Tier
This panel demonstrates secure OAuth integration, server-side API routing, and live data rendering.
Let’s build useful, high-quality software.
I’m open to software engineering opportunities where I can contribute to product quality, velocity, and long-term maintainability.