James Lussier

Senior Software Engineer with 8+ years delivering scalable backend systems and agentic AI workflows — from cloud-native platforms onboarding thousands of users to multi-agent LLM orchestration that collapsed creative production from hours to minutes. B.S. in Computer Science, University of Texas at Dallas.

Tap any line to expand it — or flip to verbose.

• Senior Software Development Engineer Laptop Freedom

  May 2025 — Apr 2026 Orlando, FL
  • ▸ Architected Rocketeer ads platform — 6× faster time-to-first-ad
    Architected Rocketeer ads platform as a scalable software system using Java Spring Boot microservices and domain-driven data modeling on PostgreSQL relational databases that onboarded thousands of non-technical entrepreneurs, delivering 6x faster time to first ad and sub-10-minute campaign launches.
  • ▸ Integrated Facebook Marketing API + Mapbox — −70% setup errors
    Integrated the Facebook Marketing API and Mapbox to enable end-to-end campaign management with geographic targeting, reducing campaign setup errors by 70% and eliminating reliance on Meta Ads Manager.
  • ▸ Built n8n multi-agent OpenAI workflow — hours to 2 minutes
    Built an n8n workflow orchestrating 5-7 OpenAI agents with AutoGen and prompt engineering, creating agentic workflows that automated ad copy and creative generation with Nano Banana text-to-image, cutting production time from hours to 2 minutes.
  • ▸ Evaluated emerging LLMs to guide AI investment strategy
    Applied LLM engineering to evaluate emerging models such as OpenClaw and NemoClaw with LangSmith, benchmarked ad quality using Claude Code and LangGraph, and defined selection criteria to guide the platform's long-term AI investment strategy toward models that improve ad relevance.
  • ▸ Built containerized CI/CD infrastructure — −80% deployment time
    Built containerized infrastructure on MCP servers and CI/CD pipelines using Docker, Terraform, and GitHub Actions, reducing deployment time by 80% and supporting weekly releases with near-zero production incidents.
  • ▸ Mentored engineers — ramp-up cut from a month to a week
    Mentored junior developers via GitHub code reviews, pair-programming sessions, and architecture guidance on Node.js and React projects, cutting new-hire ramp-up from one month to one week, lowering bug rates in shipped code, and boosting overall team velocity.
• Full Stack Software Engineer | AI Integration FranklinCovey Co

  Oct 2020 — Apr 2025 Draper, UT
  • ▸ Built backend APIs for secure, low-latency financial analytics
    Produced and optimized backend services and RESTful APIs with C#, .NET, and SQL databases, enabling high throughput, low-latency financial analytics and secure authentication, which reduced processing delays and ensured safe handling of student information.
  • ▸ Led AI-powered feature architecture for education apps
    Led full-stack development and defined AI-powered features architecture for education applications using React, Vue.js (Nuxt.js), Next.js, Python, TypeScript, and GraphQL, resulting in streamlined learning workflows and improved communication between students and instructors.
  • ▸ Shipped real-time voice streaming + transcription at scale
    Implemented multi-threaded backend services using FastAPI and WebSockets to support real-time voice streaming, speech-to-text transcription, and concurrent audio sessions at production scale, enabling live classroom environments to run smoothly without interruption.
  • ▸ Cut educator prep time 45% with generative AI tooling
    Expanded Generative AI learning tools using Anthropic LLMs, the OpenAI SDK, and Cursor, applying agent development to reduce educator preparation time by up to 45% through automated lesson planning and content summarization.
  • ▸ Built voice-enabled, multimodal AI learning assistants
    Built multimodal, voice enabled AI learning assistants that handled student onboarding, performed real time transcription of lectures, and enabled conversational tutoring using Whisper, Deepgram, LiveKit, and custom LLM pipelines, resulting in faster onboarding and higher student engagement.
  • ▸ Designed FERPA-compliant RAG + vector DB architectures
    Designed vector database architectures and RAG systems with large-scale multimodal data handling and embeddings to deliver context-aware learning insights, helping instructors make faster, more informed instructional decisions while keeping student data FERPA compliant.
  • ▸ Deployed production AI on SageMaker/Bedrock — −40% inference cost
    Deployed production AI systems on AWS SageMaker and Bedrock using Terraform across Azure and GCP, reducing inference costs by 40% while scaling to millions of monthly requests.
  • ▸ Upgraded summarization systems for 10,000+ monthly sessions
    Upgraded advanced summarization systems that generated high quality lecture notes, study guides, and session summaries, reducing instructor documentation time by 40% across 10,000+ monthly learning sessions.
• Software Developer Unlesh

  Oct 2018 — Sep 2020 Denver, CO
  • ▸ Modernized learning management and academic operations
    Engineered full-stack applications with React, Next.js, Node.js, and Python that modernized learning management and academic operations, resulting in faster course setup and reduced administrative workload for education clients.
  • ▸ Improved enrollment forecasting accuracy 15%
    Delivered custom software solutions that improved enrollment forecasting accuracy by 15%, streamlined tracking of 5,000+ learning resources and digital assets across multiple campuses, and strengthened program oversight for multi-site academic institutions.
  • ▸ Built FERPA-aligned backend services — 99.9% uptime
    Designed scalable backend services on AWS using C# and .NET Core with multi-threaded architecture to securely handle FERPA-aligned student data and deliver 99.9% uptime for learning platforms, enhancing system reliability for academic users.
  • ▸ Built responsive dashboards with Tailwind CSS
    Built responsive interfaces with Tailwind CSS that improved data-visualization clarity for academic dashboards, enabling faculty and administrators to make faster instructional decisions and reducing onboarding time for new users by 40%.
  • ▸ Established CI/CD pipelines with GitHub Actions
    Established CI/CD pipelines with GitHub Actions to automate testing, code reviews, and deployments, reducing release cycles from weeks to days and improving code quality across distributed development teams supporting education clients.
  • ▸ Translated Figma handoffs into WCAG-compliant React UI
    Collaborated with UX/UI designers through structured design reviews and Figma handoffs, translating pedagogical requirements into WCAG-compliant React interfaces that increased student and instructor engagement on learning platforms.
• Software Engineering Intern The University of Texas at Dallas

  May 2018 — Aug 2018 Richardson, TX
  • ▸ Built volunteer coordination app for 200+ church members
    Designed and developed full-stack web application to streamline volunteer coordination, reducing manual scheduling overhead and enabling 200+ church members to self-enroll in events through a centralized platform.
  • ▸ Architected MySQL DB + C# .NET API for member data
    Architected MySQL database and built C# .NET API endpoints to securely store and manage member contact information, eliminating reliance on disorganized spreadsheets and enhancing data accessibility for ministry leaders.
  • ▸ Automated volunteer routing — days to minutes
    Implemented automated email notifications with C# .NET SMTP libraries, routing volunteer applications to position managers and reducing response times from days to minutes.