AI Trainer - Support Operations

Homebase•Houston, TX

7d•Hybrid

About The Position

We're looking for an AI Trainer – Support Operations who's excited to shape how our AI agent thinks, responds, and resolves customer issues—end to end. This role owns how our customer-facing AI agent (powered by Sierra AI) operates: how it handles conversations, routes customer needs, executes workflows, and improves over time. It sits at the intersection of AI training, support operations, and workflow design—with a strong focus on measurable outcomes like containment rate, CSAT, and resolution quality. This is not a traditional support role. It's a specialist position focused on training and optimizing a production AI agent that serves real customers at scale. If you're energized by data, fluent in AI, and motivated by turning messy problems into clean, repeatable solutions—this role is for you.

Requirements

2+ years of experience in AI training, support operations, conversation design, chatbot management, or a closely related role
Hands-on experience working with AI tools—ideally building or optimizing a customer-facing chatbot, AI agent, or automated workflow
Demonstrated ability to use data to identify problems, prioritize work, and measure impact
Strong written communication skills—a significant portion of this role is writing: journeys, documentation, workflows, and prompts
Comfort with cross-functional collaboration, including working with Product and Engineering teams
A bias for action: you don't wait for perfect information to start moving
Ability to work in-office Monday–Thursday at our Houston hub (non-negotiable)

Nice To Haves

Experience with Sierra AI, LLM prompt engineering, or similar AI agent platforms
Background in support operations, QA, or workflow automation
Familiarity with tools like Salesforce, Talkdesk, Level AI, or Ask AI
Experience defining escalation logic, containment strategies, or AI performance benchmarks

Responsibilities

Build and optimize conversational journeys within Sierra AI—defining how the agent gathers context, handles branching logic, and resolves customer issues without human handoff.
Translate complex support workflows into structured, testable AI journeys that the agent can execute accurately and confidently.
Identify gaps in agent behavior through data analysis, then design, test, and iterate on solutions to close them.
Own the full lifecycle: identify → design → build → test → monitor → improve.
Analyze contact data, containment rates, and customer sentiment to identify the highest-impact areas to focus on.
Use tools like Sierra AI, Level AI, and Ask AI to surface trends, prioritize work, and measure the impact of every change you ship.
Track key metrics—containment rate, CSAT, resolution accuracy—and use them to guide decisions and communicate progress.
Partner with Product Managers, Engineers, and Support SMEs to scope and build AI capabilities that require backend access or new action tools.
Document workflows clearly in Google Docs and communicate asks to engineering and product in a way that's specific, well-reasoned, and actionable.
Proactively flag opportunities and surface data-backed recommendations to your manager and cross-functional partners—not just wait to be assigned work.
Rigorously test new journeys before they go live—catching edge cases, error states, and failure modes before customers see them.
Monitor live journeys post-launch and continuously refine based on real customer behavior.
Contribute to a culture of experimentation: form a hypothesis, build it, measure it, iterate.