<p><br><strong>Location Qualifications:</strong> This role is available for hybrid work (2 days on-site) from our office in Bucharest.<br><br></p> <p><strong>About the position</strong></p> <p>Our data platform powers customer-facing embedded analytics that thousands of users rely on daily. But we're going further: Showpad is building a Revenue Intelligence engine that transforms raw signals — CRM data, email threads, call transcripts, and content engagement — into prescriptive AI guidance. You will be a foundational architect of the unified data intelligence layer that makes this engine possible.</p> <p>You will participate in the design of our Medallion architecture, star schemas, and data governance frameworks — and critically, you will help define how structured and unstructured knowledge is modeled, connected, and retrieved. Your goal is to build a platform so robust and intuitive that other teams are empowered to build high-quality downstream models with speed and autonomy.</p> <h3><strong>What You’ll Do</strong></h3> <ul> <li>Shape how we model and connect revenue data — accounts, deals, contacts, content, and engagement signals — across both structured and <strong>semantic</strong> layers</li> <li>Design and evolve the <strong>ontology</strong> that defines how entities relate, enabling relationship-aware reasoning beyond what traditional joins can offer</li> <li>Build and maintain a <strong>Knowledge Graph</strong> that supports multi-hop inference and entity disambiguation across our Revenue Intelligence engine</li> <li>Own our <strong>retrieval architecture</strong> — choose the right approach for each use case, whether that's RAG, vector similarity, graph traversal, or structured search</li> <li>Define how we ground <strong>LLM outputs</strong> in proprietary data, ensuring AI features are accurate, explainable, and production-ready</li> <li>Contribute to our Medallion architecture, star schemas, and data catalog, keeping the foundation solid as we scale</li> <li>Ensure our embedded analytics meet the demands of multi-tenant isolation, sub-second latency, and strict SLAs for white-labeled products</li> <li>Be a go-to resource for cross-functional teams — setting standards, reviewing models, and helping others build confidently on top of the platform</li> <li>Establish and maintain data contracts, schema governance, and lineage practices that keep data trustworthy at every layer</li> <li>Collaborate with Product and Engineering to turn complex business needs into clear, pragmatic technical roadmaps.</li> </ul> <h3><strong>Our Tech Stack</strong></h3> <p class="font-claude-response-body break-words whitespace-normal">Deep in the AWS ecosystem, with a growing AI/ML and semantic layer:</p> <ul class="[li_&]:mb-0 [li_&]:mt-1 [li_&]:gap-1 [&:not(:last-child)_ul]:pb-1 [&:not(:last-child)_ol]:pb-1 list-disc flex flex-col gap-1 pl-8 mb-3"> <li class="font-claude-response-body whitespace-normal break-words pl-2"><strong>Orchestration & Transformation:</strong> dbt, Glue Jobs, AWS ECS Fargate</li> <li class="font-claude-response-body whitespace-normal break-words pl-2"><strong>Storage & Catalog:</strong> S3 Data Lake (Iceberg/Hive), Glue Data Catalog, Lake Formation</li> <li class="font-claude-response-body whitespace-normal break-words pl-2"><strong>Compute & APIs:</strong> AWS Lambda (TypeScript/Python), Athena</li> <li class="font-claude-response-body whitespace-normal break-words pl-2"><strong>Analytics & BI:</strong> ClickHouse, OpenSearch, AWS QuickSight, Luzmo</li> <li class="font-claude-response-body whitespace-normal break-words pl-2"><strong>Knowledge & Retrieval:</strong> Amazon Neptune (or equivalent graph DB), vector stores (e.g., OpenSearch kNN, pgvector), RAG pipelines, ontology frameworks (RDF/OWL or property graph models)</li> <li class="font-claude-response-body whitespace-normal break-words pl-2"><strong>Infrastructure:</strong> CDK / CloudFormation.</li> </ul> <h3><strong> What We’re Looking For</strong></h3> <ul> <li>You think in systems, not just solutions — you design for the problem two steps ahead, not just the one in front of you</li> <li>Strong foundation in data modeling: you know Kimball and dimensional modeling well, and you know when to bend the rules</li> <li>Comfortable working across both relational and semantic models — you understand ontology design and when a knowledge graph unlocks what a schema can't</li> <li>Hands-on experience with RAG pipelines: chunking strategies, embedding design, retrieval evaluation, and what makes them fail in production</li> <li>You've worked with large-scale, event-driven systems and know how to handle messy, unstructured inputs like transcripts, emails, and documents</li> <li>DevOps is second nature — CI/CD, monitoring, and DataOps practices are how you work, not things you bolt on later</li> <li>You can talk trade-offs clearly: a PM, an engineer, and a business stakeholder should all leave the same conversation understanding your recommendation</li> <li>You're pragmatic about tooling — you'll reach for the right retrieval strategy, not the most impressive one.</li> </ul> <p><strong>Company Highlights: </strong></p> <p>Welcome to the new era of revenue effectiveness. The merger of Showpad and Bigtincan is creating the first