AI Conversation Designer at DrFirst

Job description

For 25 years, DrFirst has empowered providers and patients to achieve better health through intelligent medication management. We improve healthcare workflows and help patients start and stay on therapy with end-to-end solutions that enhance prescription access, affordability, and adherence. Our solutions help 100 million patients a year and are used by more than 420,000 prescribers, 71,000 pharmacies, 270 EHRs and health information systems, and over 2,000 hospitals in the U.S. This is a great opportunity to be a part of a successful Healthcare IT company experiencing significant growth. Here you’ll get to work with some of the smartest and most interesting people around, solving unique and complex challenges in healthcare on a scale matched by few companies. If you get excited about stretching yourself in new ways, developing yourself to your fullest potential, and care about working with smart colleagues, we want to talk to you! DrFirst has invested in AI enabled service delivery, and those investments are live and operational today across both customer facing and agent facing surfaces. This role reports to a Director in the Customer Operations organization and owns the customer facing side. You will own a stable, high performing AI bot that resolves customer inquiries, the quality assurance program that keeps its answers trustworthy, and a Help Center accurate and well structured enough to make both possible. The role sits in the Customer Support organization and rests on two capabilities held by the same person. The first is hands on Zendesk technical proficiency. The AI surfaces cannot be improved without understanding how the platform works underneath them. The second is the ability to use AI writing tools to produce accurate customer facing content, and to stand behind that accuracy for a prescriber and pharmacist audience. This is not a Zendesk administration role. It is an AI performance and content role that connects platform expertise to how DrFirst products work and how prescribers and pharmacists experience support. Platform depth is what we are hiring for. Product depth is what we will build with the person we hire. The agent facing AI suggestion tools that assist support agents mid ticket, Zendesk CoPilot and Auto Assist, are a separate workstream owned within Customer Operations Program Management. This role is the technical diagnostic partner to that work, not its owner. Customer Facing AI Bot: Stability and Performance • Own the iPrescribe and DrFirst US AI Bot as a product, not just a configuration. • Monitor bot performance continuously across containment rate, escalation triggers, conversation quality, and intent coverage. • Identify patterns in bot failures and escalations, and convert them into content and configuration fixes rather than reports. • Maintain escalation logic and routing rules in close partnership with Zendesk Administration. • Define and track stability benchmarks, and flag regressions proactively before they affect the prescriber experience. Customer Facing AI Bot: Quality Assurance • Own the conversation quality assurance program for the customer facing bot, including the review sample, the scoring standard, and the review cadence. • Review bot conversations against that standard to confirm responses are accurate, on brand, and appropriate for a healthcare prescriber and pharmacist audience. • Trace quality failures to their source in intent design, content accuracy, or configuration, and route each to the fix that resolves it. • Report quality assurance findings as a trend line rather than a list of individual conversations, so recurring failure modes are visible and get fixed once. Agent Facing AI Suggestions: Technical Diagnosis • Serve as the technical resource when agent facing AI suggestions are inaccurate or unhelpful, and trace root cause to intent, content, or configuration. • Fix the causes that sit in this role scope: Help Center accuracy, content structure, and intent overlap between the customer facing bot and the agent facing surfaces. • Hand configuration and adoption causes to the owning workstream with the diagnostic evidence attached, rather than logging them as tickets. Help Center: The AI Fuel Layer • Own the AI performance view of Help Center content, and identify articles that are outdated, missing, or structured in ways that degrade answer quality on either surface. • Author and publish content changes where the trigger is AI performance, including bot failure remediation, intent coverage gaps, and restructuring existing articles for retrieval quality. • Draft and revise that content using AI writing tools approved for use at DrFirst, working within DrFirst data handling parameters. Accountability for factual accuracy, product correctness, and brand voice sits with this role, not with the tool. • Own the content structure standard that governs AI suggestion quality, and make it available to the Learning team so release driven articles meet it at publication. • Track Help Center search performance and no result rates as leading indicators of content gaps, and route release driven gaps to Learning with the supporting performance evidence attached. • Review release driven article changes for AI performance impact where an AI performance fix and a release change touch the same article. AI Performance Measurement and Reporting • Build and maintain a consistent AI performance reporting framework in partnership with the Director, Customer Operations Program Management. • Deliver regular reporting covering bot containment rate, escalation patterns, Help Center search performance, and content gap trends. • Feed AI performance data into your director operational reporting and the Customer Operations executive dashboard. • Surface trends, anomalies, and improvement opportunities proactively. Who You Will Work With • Your Director in Customer Operations: your direct manager, who sets priority and aligns AI performance work to support vision and agent experience goals. • Director, Support Delivery: your customer outcomes partner, for whom you supply the data and the fixes behind resolution quality, escalation volume, and CX. • Director, Customer Operations Program Management: your metrics partner and the owner of the agent facing AI suggestion workstream. • Director, Customer Training and Knowledge Management: your shared Help Center partner, divided by what triggers the work. • Zendesk Administration: your platform partner, working in close coordination on configuration, routing rules, and platform changes. • VP, Customer Operations: strategic alignment, because AI enabled service delivery is a VP level OKR and this role is the operational engine behind it. What Success Looks Like in 12 Months • The AI bots hold a stable, improving containment rate against documented performance benchmarks. • Bot conversation quality is reviewed on a defined standard and cadence, and recurring failure modes are fixed at source rather than re reviewed each cycle. • Agent facing AI suggestion issues raised by agents are diagnosed to root cause quickly, with content side causes fixed rather than logged. • Zendesk Administration trusts this role as a Zendesk AI subject matter expert and a reliable platform partner. • Your director and the Director, Customer Operations Program Management have the data they need without having to chase it. • Help Center articles you author hold up under review, and AI assisted drafting reaches publication without accuracy rework. • Release driven and AI performance driven Help Center updates land without collision, with neither team blocked waiting on the other. • The escalation rate from AI to human declines quarter over quarter. Required • Hands on Zendesk technical proficiency. This is the single most important qualification for the role and the one we screen hardest on. You show working knowledge of Help Center and Guide structure, macros, triggers and automations, views and routing, and Zendesk reporting, and you are prepared to walk through Zendesk work you have personally built or configured. • Demonstrated ability to use AI writing tools to produce accurate customer facing content. This role drafts and revises Help Center articles with AI assistance and is accountable for factual accuracy, product correctness, and brand voice in the published result. You can describe how you verify AI generated content before it reaches a prescriber or pharmacist audience. • Conversation and intent design capability. You structure intents, coverage, and dialogue flows so that a bot resolves rather than deflects. Formal conversation design training is welcome but not required if you show the work. • Hands on familiarity with at least one Zendesk AI surface: a customer facing AI bot, agent facing AI suggestions such as CoPilot or Auto Assist, or Help Center content management. • Analytical mindset. You work comfortably with platform data, identify performance patterns, and draw conclusions without being directed to look for specific things. • Strong written communication. Help Center articles, diagnostic write ups, and performance reporting are the core outputs of this role. • Comfort working cross functionally without formal authority, influencing through platform expertise rather than hierarchy. Strongly Preferred • Working knowledge of prescriber or pharmacy workflows, or prior support experience in a healthcare software environment. • Experience identifying ticket drivers, escalation patterns, or bot conversation gaps in a support environment. • Familiarity with Zendesk reporting and analytics, including the ability to build or interpret dashboards that reflect AI performance metrics. • Self directed learning in AI tools applied to customer service. You went looking for this knowledge without being asked. • 90% desk work • 10% remote travel to client sites, conferences, or internal team meetings #LI-GF1 #LI-Remote • Competitive compensation, with a base salary of $60,000 - $75,000 (Exact compensation may vary based on skills and experience) • Eligible for a Discretionary Performance-based bonus • Medical, dental, and vision insurance • 401K eligible after 3 months of employment, with 50% company match up to first 5% of salary contributed to the plan with a 3-year vesting schedule • HSA for eligible employees enrolled in the HDHP, with a generous company contribution up to $500 for individual coverage and $1000 for family coverage per year • 100% company paid short and long-term disability, AD&D, and group life insurance • Accrued annual paid time off (PTO) of 18 days for the first 3 years of service, increasing thereafter and 7 paid holiday days • Employee Assistance Program • Continuing Education funds up to $1500 annually for eligible programs after 1 year of service • Voluntary benefits including FSA, Hospital indemnity, Accident and Critical Illness insurances DrFirst is committed to being a Remote-First company, creating a dynamic and flexible workplace where everyone thrives, no matter where they log in from. Check out our approach to remote work: drfirst.com. Our recruitment process at DrFirst is straightforward and secure. You will only be contacted by our recruitment team through an official @drfirst.com email address. We will never ask you for payment or sensitive personal information, such as your social security number or banking details, at any stage of the hiring process. Additionally, we will not request that you purchase equipment or accept e-checks or checks for deposit. If you encounter any communications claiming to be from DrFirst that seem suspicious, please contact our recruitment team directly at recruiter@drfirst.com to verify the message’s authenticity. Your security is important to us! Learn more about our benefits and professional development opportunities: drfirst.com.

Requirements

Must-haves

Customer Facing AI Bot: Stability and Performance
Own the iPrescribe and DrFirst US AI Bot as a product, not just a configuration.
Monitor bot performance continuously across containment rate, escalation triggers, conversation quality, and intent coverage.
Identify patterns in bot failures and escalations, and convert them into content and configuration fixes rather than reports.
Maintain escalation logic and routing rules in close partnership with Zendesk Administration.
Define and track stability benchmarks, and flag regressions proactively before they affect the prescriber experience.
Customer Facing AI Bot: Quality Assurance
Own the conversation quality assurance program for the customer facing bot, including the review sample, the scoring standard, and the review cadence.
Review bot conversations against that standard to confirm responses are accurate, on brand, and appropriate for a healthcare prescriber and pharmacist audience.
Trace quality failures to their source in intent design, content accuracy, or configuration, and route each to the fix that resolves it.
Report quality assurance findings as a trend line rather than a list of individual conversations, so recurring failure modes are visible and get fixed once.
Agent Facing AI Suggestions: Technical Diagnosis
Serve as the technical resource when agent facing AI suggestions are inaccurate or unhelpful, and trace root cause to intent, content, or configuration.
Fix the causes that sit in this role scope: Help Center accuracy, content structure, and intent overlap between the customer facing bot and the agent facing surfaces.
Hand configuration and adoption causes to the owning workstream with the diagnostic evidence attached, rather than logging them as tickets.
Help Center: The AI Fuel Layer
Own the AI performance view of Help Center content, and identify articles that are outdated, missing, or structured in ways that degrade answer quality on either surface.
Author and publish content changes where the trigger is AI performance, including bot failure remediation, intent coverage gaps, and restructuring existing articles for retrieval quality.
Own the content structure standard that governs AI suggestion quality, and make it available to the Learning team so release driven articles meet it at publication.
Track Help Center search performance and no result rates as leading indicators of content gaps, and route release driven gaps to Learning with the supporting performance evidence attached.
Review release driven article changes for AI performance impact where an AI performance fix and a release change touch the same article.
AI Performance Measurement and Reporting

Nice-to-haves

Strongly Preferred
Eligible for a Discretionary Performance-based bonus

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

Platform: icims
Effort: medium

Benefits

Benefits detected in the listing.

Dental

Vision

401k

Pto

Paid Time Off

Professional Development