PlayOn Sports processes over 250,000 live high school sports games a year across NFHS Network, GoFan, and MaxPreps. We’re using AI and computer vision to turn those streams into real-time scores, player stats, automated highlights, and interactive fan experiences. As a Senior Engineer on the Streaming Intelligence team, you’ll build the human-in-the-loop and fan-in-the-loop interfaces that connect our computer vision pipeline to the people who use it — internal operators reviewing AI-generated stats and millions of fans engaging with AI-powered experiences across our three consumer brands.
In this role, you’ll own the interactive layer between AI and users: annotation review tools, real-time stat overlays, correction workflows, and fan-facing features that run at scale across web and mobile. You’ll work primarily in Python, ship features end-to-end, and think AI-forward — not just consuming model outputs, but designing interfaces and services that make AI systems better through human feedback and fan interaction.
The ideal candidate is a builder. You’re energized by shipping, comfortable with ambiguity, and excited about working at the intersection of AI and product. You don’t wait for a fully baked spec — you break down loosely defined problems and start delivering.
The outcomes you’ll deliver
- Production annotation review interface: Ship the first human-in-the-loop interface for the computer vision stats pipeline, enabling internal operators to review, correct, and approve AI-generated statistics in real time. Target: production-ready within six months.
- Fan-facing AI feature: Deliver at least one AI-powered fan experience — real-time stat overlays, interactive highlights, or personalized content — to one of our consumer brands (NFHS Network, GoFan, or MaxPreps). Target: live within nine months.
- Reusable AI development patterns: Establish the team’s standard UI component library and Python service templates for AI-forward development, enabling faster iteration on future human-in-the-loop and fan-in-the-loop features.
- Correction workflow at scale: Build feedback loops that capture human corrections and fan interactions and route them back into the AI pipeline, measurably improving model accuracy over time.
- Cross-brand consistency: Deliver interactive AI features that work reliably across NFHS Network, GoFan, and MaxPreps, adapting to each brand’s UX patterns while sharing a common service layer.
In this role, you can expect to
- Build and ship human-in-the-loop interfaces that enable internal operators to review, correct, and approve AI-generated sports statistics in real time across thousands of concurrent live streams.
- Design and develop fan-in-the-loop experiences — interactive highlights, live stat overlays, and engagement features — across NFHS Network, GoFan, and MaxPreps.
- Own features end-to-end: from data modeling and API design through frontend implementation and production deployment. You’ll ship, not just spec.
- Integrate AI/ML model outputs (computer vision, LLMs) into production applications, building the service layer between models and users.
- Develop reusable Python service templates and UI component patterns that establish the team’s standard for AI-forward development.
- Collaborate with computer vision engineers, product managers, and data teams to translate pipeline outputs into intuitive, performant user experiences.
- Contribute to system design and architecture decisions within the streaming intelligence program. Participate in code reviews, design reviews, and technical documentation.
- Help evaluate and integrate third-party tools and vendor APIs (annotation platforms, model serving infrastructure) as the platform scales.
To thrive in this role, you have
- 3+ years of professional software engineering experience with strong Python skills and a track record of building production web applications and APIs.
- Builder mentality — you’ve shipped end-to-end features from concept to production, not just maintained existing systems. You bias toward action and iterate quickly.
- Experience integrating AI/ML models into user-facing applications (LLM APIs, computer vision pipelines, or similar). You understand the practical challenges of making model outputs useful to real people.
- Solid fundamentals in API design, data modeling, and service architecture. You write clean, testable code and care about the systems you leave behind.
- Familiarity with cloud infrastructure (AWS EKS, S3) and modern data tooling (Snowflake, Kafka, or similar).
- Strong communicator who works well across disciplines — you can talk to a product manager about user flows and a data scientist about model outputs in the same afternoon.
- Bonus: frontend experience with React/TypeScript (especially interactive annotation or dashboard UIs), familiarity with sports data or video analytics, experience with annotation tooling (Roboflow, CVAT, Label Studio), or interest in developer experience and internal tooling.