About
Jun Heider is a solutions architect and systems builder with 20+ years designing and delivering production software for media, enterprise, and government organizations. He designs AI-enabled and media-workflow-focused systems, and works with the teams who operate them afterward.
His current work is applying AI to build software and solutions in the most efficient way available: AI agents, agentic tooling, and Model Context Protocol servers, most of it in OTT and streaming. The judgment matters as much as the tooling, so he weighs whether AI belongs in a given task rather than assuming it does, and works trust but verify, because these systems can mislead and hallucinate. Node and TypeScript are the deeper stack, with Python and AWS carrying the agent work, alongside authored Claude Skills, MCP servers for domain tooling, and GitHub Copilot automation. He has also built multi-agent platforms, evaluation harnesses, and retrieval-augmented generation in secured environments.
That work sits on a career in streaming and media infrastructure: transcoding pipelines, CDN architecture, live delivery, and playback performance for organizations ranging from startups to Peacock and Dolby. He was Director of Media Streaming Solutions at Dolby Laboratories and Chief Technology Officer at RealEyes Media, a streaming technology consultancy, where he spent fifteen years working directly with customers.
He teaches as well as builds: work in progress on a seven-module course covering AI across the software development lifecycle, subject matter expert work with the Streaming Video Technology Alliance, and talks at Demuxed, NAB Streaming Summit, Streaming Media, and Streaming Tech Sweden.
Current
Principal, Ajnunna Systems: AI systems and streaming consulting for product teams
Subject Matter Expert and Committee Co-Chair, Streaming Video Technology Alliance (SVTA): co-chair of the Industry References Committee, contributor to the Metadata and Measurement/QoE working groups
Professional Experience
Director of Media Streaming Solutions at Dolby Laboratories, Chief Technology Officer at RealEyes Media. Jun has led critical media orchestration for the Peacock OTT service, and managed product development and DevOps infrastructure.
Selected Work
copilot-assigner-worker: Cloudflare Worker that assigns GitHub issues to the Copilot coding agent over OAuth, so Actions workflows can hand work to the agent.
shaka-packager-mcp-server: Model Context Protocol server exposing Shaka Packager to AI agents, so media packaging can be driven directly from an agent session.
vse-cmcd-monitor: streaming anomaly detection over Common Media Client Data, built in the open with the AI pair-programming session logs committed alongside the code.
More at github.com/coderjun.
Education
Bachelor of Science in Computer Networking from Regis University.
Skills
AI and development practice: agentic development workflows, AI-assisted software delivery, Claude Skills and Model Context Protocol, GitHub Copilot and GitHub Actions automation, multi-agent systems, LLM evaluation and golden datasets, retrieval-augmented generation, Node and TypeScript, Python and FastAPI, AWS, Docker and Kubernetes, CI/CD and test automation.
Streaming and media: video transcoding and packaging, HLS and DASH delivery, CDN architecture, FFmpeg, live streaming, playback optimization, DRM, OTT application lifecycle.
Publications and Conferences
Co-author, Professional Adobe Flex 3 (Wrox/Wiley).
Project lead, SVTA1023-1: Content Metadata Landscape Revision 1.0, Metadata Working Group. Contributor, SVTA2070, Standardized Error Codes, Measurement and QoE Working Group.
Guest author, "Content Metadata Reliability Engineering" for Wowza.
Presented at Demuxed, NAB Streaming Summit, Streaming Media, and Streaming Tech Sweden. The preview of "AI for the Streaming SDLC" is live at SVTA University; the full seven-module course is in development.