Search Engineer/Tech Lead

Working Relationships Collaborates with Enterprise Architecture, Digital Experience product owners, AEM/front-end engineering, commerce and product-data teams, AI platform teams, cloud operations, cybersecurity, privacy/legal, analytics, content owners, and implementation partners. The role delivers secure, grounded, and supportable conversational search using Lucidworks Fusion or a comparable enterprise search platform, with an initial single-turn answer experience followed by multi-turn conversational search.

Duration: 6 months, possibility of extension

Job Description:
Summary

• Lead the design, build, tuning, and operation of the search layer for client.com.
• The role delivers secure, grounded, and supportable conversational search using Lucidworks Fusion or a comparable enterprise search platform, with an initial single-turn answer experience followed by multi-turn conversational search.
• Translate customer and business needs into scalable index and query pipelines across web, content, commerce, and AI platforms.
• This contract role is an alternative to external professional services and requires hands-on technical delivery as well as solution leadership.

Responsibilities
Search Platform and Retrieval Engineering

• Design and implement collections, schemas, connectors, index/query pipelines, query profiles, and REST API integrations in Lucidworks Fusion or an equivalent platform.
• Build and tune lexical, semantic, vector, and neural hybrid retrieval, including embeddings, blend weights, thresholds, boosting, filters, facets, synonyms, and fallback behavior.
• Use signals and behavioral data to improve relevance, recommendations, personalization, and popular or successful results.
• Create relevance benchmarks, automated regression tests, citation and grounding tests, latency Clients, and per-query cost measures.
• Troubleshoot ingestion, indexing, Solr, Kubernetes/AWS, APIs, pipelines, model endpoints, and front-end integrations.

Conversational Search and Generative AI

• Engineer single-turn and multi-turn search flows with intent recognition, entity extraction, query rewriting, clarification, and session context.
• Implement grounded RAG that retrieves approved sources, cites evidence, and suppresses or falls back when evidence or confidence is insufficient.
• Integrate with enterprise LLM services and client s AI Gateway using prompt controls, model configuration, rate limits, error handling, latency budgets, and cost governance.
• Tune experiences for product numbers, specifications, availability, certificates, manuals, application notes, and related support content.
• Apply guardrails for transactional or product-SKU queries, low-confidence grounding, and zero-result scenarios.

Integration, Delivery, and Operations

• Partner with AEM and front-end engineers to deliver accessible search results, conversational answers, follow-up suggestions, and facets aligned with the client s Design System.
• Manage Fusion configuration as code through GitLab, peer review, automated testing, and CI/CD practices.
• Instrument click-through, zero-result, no-click, reformulation, abandonment, task completion, latency, and pipeline-health metrics.
• Produce architecture diagrams, deployment documentation, runbooks, configuration standards, and knowledge-transfer materials.
• Work with security, privacy, legal, architecture, product, content, and business teams on access controls, data handling, AI guardrails, and release readiness.
• Participate in client's AI-DLC and Agile delivery cadence.

Required Education, Experience, and Skills

• Bachelor's degree in Computer Science, Information Systems, Engineering, Data Science, or a related discipline, or equivalent practical experience.
• 5+ years in search engineering, information retrieval, or enterprise application development, with substantial experience in Lucidworks Fusion or comparable platforms such as Solr/Lucene, Elasticsearch/OpenSearch, Algolia, Vespa, or Azure AI Search.
• Hands-on knowledge of schemas, analyzers, tokenization, synonyms, faceting, boosting, filtering, relevance scoring, query debugging, connectors, signals, custom pipeline stages, and REST APIs.
• Practical experience with semantic search, embeddings, approximate nearest-neighbor retrieval, neural hybrid ranking, relevance evaluation, RAG, grounding, prompt design, citations, guardrails, and fallback patterns.
• Experience with multi-turn context, session design, intent classification, and entity extraction.
• Experience deploying or operating search workloads in AWS and Kubernetes, including observability and production troubleshooting.
• Proficiency in Java and/or Python, JSON, HTTP APIs, Git, automated testing, and CI/CD.
• Ability to convert customer journeys and business requirements into technical designs, backlog items, acceptance criteria, and measurable outcomes.
• Clear communication with technical and nontechnical stakeholders; able to work independently, prioritize competing work, resolve ambiguity, and transfer knowledge.

Preferred Qualifications

• Lucidworks Fusion 5.x implementation, upgrade, or administration experience in a self-hosted environment.
• Experience with Fusion AI, RAY, Learning-to-Rank, Relevance Workbench, Analytics Studio, Commerce Studio, A/B testing, or equivalent capabilities.
• Integration experience with AEM as a Cloud Service, commerce, product catalogs, digital assets, or certificate/document services.
• Knowledge of Redis or another session store; multilingual search; part-number/SKU handling; permissions-aware retrieval; and structured/unstructured content blending.
• Experience in compliance-sensitive or document-intensive environments, plus familiarity with accessibility, privacy, secure AI development, responsible AI, and production monitoring.
• Relevant search, cloud, or AI certifications.

Key Deliverables and Outcomes

• Production-ready architecture for lexical, semantic, hybrid, and conversational search.
• Grounded conversational pipelines that retain context, identify intent and entities, cite evidence, and use appropriate fallback behavior.
• Relevance benchmarks and regression tests for priority customer journeys and product-search patterns.
• Operational dashboards and alerts for quality, adoption, latency, failures, and model or gateway dependencies.
• Runbooks, deployment documentation, standards, and knowledge transfer that enable sustainable ownership.

Core Competencies

• Search engineering | Relevance optimization | Conversational AI | RAG and grounding | Systems integration | Production operations | Analytical problem solving | Cross-functional collaboration | Technical leadership

Working Relationships

• Collaborates with Enterprise Architecture, Digital Experience product owners, AEM/front-end engineering, commerce and product-data teams, AI platform teams, cloud operations, cybersecurity, privacy/legal, analytics, content owners, and implementation partners.

Success Measures

Measure	Expected Outcome
Search relevance	Improved judged relevance for priority queries without unacceptable latency or regression.
Conversational quality	Users complete multi-turn journeys with retained context and evidence-backed responses.
Trust and safety	Answers use approved sources, include citations, are monitored, and are suppressed when confidence is inadequate.
Customer outcomes	Reduced zero-result, reformulation, and abandonment rates; improved click-through and task completion.
Operational readiness	Pipelines, integrations, and model dependencies are observable, supportable, documented, and recoverable.