Justin Joyce

Justin Joyce combines experience in sales operations and machine learning to build agent-assisted analytical support for go-to-market teams. In mid-2026, he was principal sales operations and strategy manager at Cloudflare, supporting lead-generation and post-sale customer-experience teams within revenue operations. His work connects business knowledge with customer data so employees can prepare for conversations and understand performance, while operations specialists gain more time for strategy.

From sales operations to machine learning

Joyce began in sales and sales operations, then spent approximately seven years working in machine learning at Grainger. He wanted to develop prescriptive analysis: using predictions to help a business decide what action to take next. Returning to sales operations at Cloudflare allowed him to combine that analytical experience with his understanding of commercial work.

By 2021, his authored-writing profile identified him as a senior data scientist at Grainger. His weighted ensemble forecasting tutorial illustrates his interest in practical decision-support methods. He implemented a method developed by other researchers that uses time-series characteristics to determine how much weight to give different forecasting models, rather than choosing one model for every series.

The example used commodity-price data and artificially generated training series. Joyce credited the method’s researchers, linked supporting material in his Learnings repository, and reported limited success: the ensemble performed best on three of ten test series. The tutorial gives readers an implementation to explore alongside its observed limitations.

Making expertise available when employees need it

Joyce distinguishes two recurring obstacles to customer-facing work. The context gap is the effort required to gather the right information when moving between prospecting, adoption, and customer-satisfaction conversations. The expert gap separates an experienced colleague’s judgment from that of someone still learning how to assess a customer’s needs or handle an objection. Effective assistance therefore needs both relevant data and guidance about how to use it.

His approach to agent-assisted operations combines three ways of meeting those needs:

• Reusable business skills: Role-specific skill files connect business definitions, data relationships, and recurring analytical questions. Employees can investigate changes in an opportunity’s amount or closing date without routing every request through someone who can write SQL. Operations colleagues also reuse that knowledge when building applications. Joyce’s aim is to reduce routine analysis so the team can spend more time on strategy and enablement.

• Proactively delivered insights: Weekly summaries explain progress against goals, trends, strong performance, and areas requiring attention. Joyce favors bringing this information directly to employees because dashboard use varies. His team prepares the relevant filters and aggregations by time period, management group, region, and metric before agents analyze the data. The reporting workflow separates drafting, factual checking, and presentation: one agent retrieves data through Model Context Protocol connections and drafts the analysis, another checks factual consistency, and a third uses example messages to balance risks and opportunities. The team can inspect each model call’s inputs and responses; Joyce describes reviewing individual runs over two to three months during development.

• Curated self-service assistance: Joyce and his colleagues use Cloudflare OS, an internal agent workspace, for forecast briefs, quarterly business-review decks, account planning, and renewal preparation. A renewal workflow can examine customer usage to help an employee prepare an adoption or upsell conversation. The team-built workspace runs on Cloudflare Workers and Durable Objects, combining skills, Model Context Protocol connections, and an AI gateway to bring business guidance and connected data into the same working environment.

These approaches accommodate different habits: some employees ask operations a question, some respond to regular summaries, and others prefer to investigate independently. Joyce treats skill curation as central to making all three useful. Operations and go-to-market colleagues submit and review skills centrally, sharing expert knowledge while limiting overlapping instructions. Internal feedback helps the team assess whether workflows actually help employees and identify what needs improvement.

Extending assistance into business systems

In 2026, deeper meeting integrations, quoting, approvals, and Salesforce updates remained under development. Joyce envisaged attaching prepared customer materials to meetings and using checked workflows for consequential system updates. His position is that wider adoption should bring a more deliberate approach to shared business information, keeping systems aligned while preserving employees’ flexibility for ad hoc work.