Forward Deployed AI Engineer

Omnilex seeks a Mid-level Forward Deployed AI Engineer in Zürich to lead technical onboarding, optimize retrieval systems, and build AI solutions for legal professionals in a hybrid role.

This position focuses on deploying and refining AI-driven legal research tools for law firms across Europe. The engineer will bridge the gap between prototype technology and stable, real-world usage, ensuring that the platform meets the strict accuracy and traceability demands of legal workflows.

Responsibilities

• Manage the technical onboarding process for new clients, which involves processing documents, creating indexes, mapping metadata, and executing validation procedures.
• Refine retrieval and reranking mechanisms to align with specific practice areas, internal taxonomies, and relevance standards defined by the customer.
• Implement custom user experience adjustments and workflow modifications, such as templates, jurisdiction presets, citation formatting, and permission-aware search results.
• Optimize prompting strategies and context handling to satisfy rigorous requirements for grounding, traceability, citation styles, and fallback behaviors.
• Establish and maintain guardrails for AI outputs, including provenance tracking, source-grounded generation, and risk-aware uncertainty patterns tailored for legal contexts.
• Develop targeted evaluation sets for each client, featuring gold questions, acceptance criteria, and critical failure scenarios.
• Conduct rapid failure analysis and deploy improvements like chunking adjustments, deduplication, reranker tweaks, and caching strategies.
• Monitor response times and operational costs by implementing batching, caching, early exits, and practical fallback paths.
• Track quality metrics and usage trends, translating client feedback into measurable fixes and clear acceptance tests.
• Collaborate with Customer Success teams and legal experts to transform client pain points into engineering tasks.
• Create deployment playbooks and integration recipes to turn customer solutions into repeatable patterns.

Requirements

• Proven experience in building or adapting search and retrieval systems for production environments, including hybrid retrieval, reranking, and query understanding.
• Demonstrated ability to transition LLM features from prototype stages to stable, real-world applications.
• Proficiency in TypeScript and Node.js development.
• Hands-on experience with at least one of the following: Azure AI Search, pgvector/PostgreSQL, OpenSearch, or Elasticsearch.
• Strong engineering judgment, including debugging capabilities, performance tuning, edge-case handling, and operational thinking.
• Comfortable working directly with customers, conducting deep technical sessions, explaining trade-offs, and producing clear written documentation.
• Fluent English language skills.
• Availability for full-time employment.
• Willingness to work in a hybrid setup, requiring at least two days per week on-site in Zurich.

Nice to have

• Proficiency in German.
• Experience integrating customer document sources and pipelines, including connectors, ETL processes, and access controls.
• Familiarity with lightweight evaluation processes, such as human labeling loops, basic agreement checks, and simple dashboards.
• Knowledge of sparse and dense retrieval approaches, including BM25 variants.
• Experience running and operating services, with Docker being a plus.
• Familiarity with Azure, NestJS, or Next.js.
• Exposure to Swiss, German, or US legal systems.

What the company offers

• Tangible customer impact: work directly affects daily trust and adoption inside legal teams.
• High ownership: run deployments end-to-end and help define reusable solution patterns.
• Fast feedback loops: see real failure modes early and influence product direction with evidence.
• Compensation: CHF 8’000–12’000 per month + ESOP, depending on experience and skills.

About the company

Omnilex is a young, dynamic team of 15 professionals with roots at ETH Zurich. The company has scaled from an early MVP to a product used daily by thousands of legal professionals in Switzerland, Germany, and Liechtenstein, and is now scaling across Europe. The company collaborates closely with clients and iterates at a market-leading pace. The company stands out by handling unique challenges, including a combination of external data, customer-internal data, and innovative AI-first legal commentaries.

• The company collaborates closely with clients and iterates at a market-leading pace.
• The company stands out by handling unique challenges, including a combination of external data, customer-internal data, and innovative AI-first legal commentaries.