Principal Applied AI Engineer - Entity Agents
ZoomInfo · Remote
About this role
## ZoomInfo — Senior Applied AI Engineer, Entity Agents ### About the role ZoomInfo’s Core Data team builds the company and contact data behind the Go-To-Market Intelligence Platform—hundreds of millions of records resolved to the right entity, kept accurate, and delivered to 35,000+ customers. This role helps move from deterministic pipelines to **AI-native data management** using **entity agents** that pursue evidence, decide attributes, and record why. You’ll build and own the production machinery that runs these agents at scale, proves versions before shipping, and enables researchers and product managers to operate without an engineer in the loop. ### What you’ll do - **Run the benchmark and golden-gate system**: schedule judge runs, golden-gate results, scorecards, and cost-per-record; enforce gating with variance/drift tracking. - **Complete the entity-agent pipeline**: build missing stages (e.g., enricher fan-out, location sets, hierarchy batch files), migrate classifiers to the shared runner, and measure throughput + unit cost via instrumentation. - **Build evidence adapters**: unify registries, live web, mail-tenant signals, and external research vendors behind a single envelope with replayable vendor storage, measured lift, kill switches, and swappable vendors. - **Deploy onto the ZoomInfo Agentic Platform**: carry company/contact agents through the production plan, build ingest adapters (envelope → match service + bulk intake), and maintain rollback paths per version. - **Make the loop self-service**: build review queue, tagging tools, blast-radius viewer, and batch runners for researchers and PMs; measure review yield and golden-record production. - **Keep agent surfaces alive**: own health/deployment for the company classifier, contact classifier, agents hub, and judge platform. - **Instrument everything**: cost per call/row/stage, quality per version, and alerts before humans notice. ### What you’ll bring - **5+ years** software engineering, including **2+ years shipping LLM-based systems to production** (agent loops, tool use, retrieval, structured output, and evaluation harnesses). - Strong **Python and TypeScript**, comfortable across API layers, batch runners, and internal web UI. - Experience building **platform components others operate** (adapters, runners, queues, internal tools with replay/idempotency/observability). - Working knowledge of **data pipelines and entity resolution** (ingestion → matching → serving; understanding costs of false merges vs missed matches). - Proven ability to take prototypes to production while preserving what the prototype proved (benchmarks/test sets as the contract). - **Instrumentation by default**: you know cost and latency budgets. - Strong bias for action and tenacity to unblock distributed work. ### Preferred - Experience with **B2B company/contact data**, identity/registry data, or entity resolution at scale. - Experience with **Claude or comparable frontier models**, agent frameworks, and eval harnesses. - Experience deploying agents onto an internal agent platform/orchestration layer. - Background in an **AI-native startup**, applied AI team, or data infrastructure company. ### Who you are - **A builder who ships** (prefers working adapters over design docs). - **An evaluator by instinct** (no releases without test sets, thresholds, and cost numbers). - **A platform thinker** (designs for others to run it and for vendor failure). - **Comfortable in the data** (can expl
Listing freshness
CronJobs last confirmed this listing 1h ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.