About the Team

Launched in 2021, Fanatics Betting and Gaming is the online and retail sports betting subsidiary of Fanatics, a global digital sports platform. The Fanatics Sportsbook is available to 95% of the addressable online sports bettor market in the U.S. Fanatics Casino is currently available online in Michigan, New Jersey, Pennsylvania and West Virginia. Fanatics Betting and Gaming operates twenty-two retail sports betting locations, including the only sportsbook inside an NFL stadium at Northwest Stadium. Fanatics Betting and Gaming is headquartered in New York with offices in Denver, Leeds and Dublin.

• Maintain and report the key metrics for our customer-facing AI agent — containment, resolution, deflection, escalation rate, repeat contact, and customer satisfaction at a case and subtype level.
• Be part of the team that stands up measurement for each new AI surface as it launches, so no surface goes live without a definition of success and the instrumentation to track it.
• Build and maintain the BI views and dashboards your team and Product partners use to read AI performance, and keep them accurate as products, policies, and promotions change.
• Write the data requirements — event definitions, tagging, table structures, and data quality checks — the metrics depend on, and partner with Data to get them built.
• Flag it early when a number stops being trustworthy. Knowing when your own data is wrong matters as much as reporting it.

• Run deep dives at the case type and subtype level to find the stall points inside the AI interaction — where customers rephrase the same question, loop, go silent, abandon, or ask for a human, and what preceded it.
• Quantify each stall point so it can be ranked: how often it happens, which subtypes and markets it concentrates in, and what it costs in escalations, repeat contacts, and resolution loss.
• Pull together signals from more than one place — conversation data, interaction quality scores, customer and associate feedback, behavioral data, and operational metrics — into one view of what's actually driving the number.
• Distinguish issues worth fixing from noise, and be candid about which is which.

• Compare how live agents resolve a contact type to how the AI agent handles the same contacts, and identify what agents are doing that the AI cannot do yet — the lookups they run, the tools they touch, the actions they take.
• Write up each gap as a clear, specific finding: what the AI would need to close it, and the containment or resolution upside if it did.
• Pay attention to the escalation boundary. Where the AI hands off, look at whether it had to.

• Sort root causes well enough to route them to the right owner and the right lever — grounding and knowledge content, prompt and instruction design, tone, tool access, handoff timing, or underlying policy ambiguity.
• Where the AI is already doing the right thing badly, describe the change specifically: what the grounding should say, where the tone misses, which instruction is producing the wrong behavior.
• Work with content, grounding, and engineering partners to get your finding specified tightly enough to be built and tested.
• Run recurring performance reviews on the case types you own, and keep a ranked, evidence-backed list of the opportunities in them.

• Turn a pile of conversations and queries into a clear, evidence-backed narrative — the pattern, the representative examples, the size of the problem, and the specific ask.
• Quantify the impact of a proposed fix — contacts avoided, resolution lift, handle time, cost, customer experience — with assumptions you can walk someone through.
• Package findings into business cases and recommendations a Product partner can act on without a translation layer, and bring them into intake with your manager.
• Follow what you handed off through to delivery, then measure whether it did what you said it would.
• Report back on what shipped, what moved, and what didn't, so the next prioritization conversation starts from evidence.

• Work closely with Data, Product, Quality, and CX, and build the relationships that make your findings land.
• Translate your analysis into clear language for technical and non-technical audiences alike.
• Write up findings and recommendations clearly, and contribute to the leadership-facing materials your team produces.
• Share what you learn. Much of what you find will be new to the people around you, and writing it down so the team can use it is part of the work.

• 2+ years in business analytics, data analysis, business operations, customer experience analytics, or a closely related role.
• Strong SQL. You can independently pull, join, and validate data across large datasets without waiting on someone else to write the query.
• Experience owning the recurring reporting and analysis behind a metric, and being accountable for the quality of it.
• Experience building BI views and dashboards (Tableau, Looker, Power BI, Sigma, or similar) that other people actually use to make decisions.
• Ability to move from "the number moved" to "here's the specific root cause, here's roughly what it's worth, and here's what I'd fix." Diagnostic instinct matters more here than modeling sophistication.
• Comfort working directly in unstructured conversation data — reading transcripts at volume, spotting the pattern, and then proving it quantitatively rather than stopping at anecdote.
• Experience writing up findings and recommendations for partners who don't report to you, and seeing some of them acted on.
• Conceptual understanding of how LLM-powered systems behave — prompting, retrieval-augmented generation (RAG), knowledge grounding, and evaluation — enough to tell a grounding problem from a prompt problem from a tool gap.
• Clear written and verbal communication. You can write a one-pager someone reads once and understands.
• High intellectual curiosity and an AI-first approach to your own work.

• Experience analyzing conversational data from chatbots, voice systems, or agent-assist tools in production.
• Comfort with Snowflake, Databricks, or a comparable cloud data platform, and with dbt or similar transformation tooling.
• Python or R for analysis beyond what SQL handles cleanly.
• Background in customer experience, contact center, or support operations analytics — particularly anything that involved understanding how human agents actually resolve contacts.
• Hands-on experience writing or editing knowledge content, groundings, or prompts for an AI system.
• Experience in sports betting, gaming, fintech, payments, or another high-growth, regulated consumer industry.
• Experience partnering with ML or Data Science teams on evaluation frameworks, A/B testing, or experiment design.
• Familiarity with knowledge base architecture and how content quality drives AI answer quality.

Depending on the role, your interview and onboarding experience may include in-person components, such as onsite interviews or Launching into Better: LIVE—a multi-day cultural immersion in New York City for full-time, non-seasonal hires. These sessions are designed to build connection and bring our culture to life, though specific travel and participation requirements will be confirmed based on your role and location. Your recruiter will provide clear guidance at each stage of the process.

For information about our benefits, please visit https://benefitsatfanatics.com/