> Markdown version of [/jobs/ext/2608470-staff-software-engineer](https://www.wearedevelopers.com/jobs/ext/2608470-staff-software-engineer). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Staff Software Engineer - **Company:** IRONCLAD, LLC - **Location:** San Francisco, CA, United States - **Experience:** Expert - **Salary:** $220,000.0 - $270,000.0 - **Contract:** Permanent contract - **Skills:** Artificial Intelligence, Automated Storage and Retrieval Systems, Content Analysis, Elasticsearch, Google Docs, Information Retrieval, Recommender Systems, Apache Solr, Large Language Models, Machine Learning Operations, Natural Language Understanding - **Published:** August 24, 2026 - **Apply:** https://www.dice.com/job-detail/0210ee04-4e7b-432e-bb1a-d24bd4c93d14 ## About the Role * 10+ years building production systems, with a substantial portion in search, information retrieval, content understanding, or recommendation systems at meaningful scale. * Demonstrated depth in one of: learned/hybrid retrieval (lexical + vector + reranking), query understanding/NLU pipelines, or production LLM agent systems - ideally more than one. * Experience with search frameworks (Elasticsearch or equivalent - Solr, Vespa, OpenSearch; embedding search) in production, including relevance tuning and reranking. * Fluency with modern LLM APIs and multi-provider orchestration (Anthropic, OpenAI, Google) - reasoning about token budgets, provider-specific tool-calling semantics, and prompt-caching trade-offs. * Experience building eval-driven workflows - offline benchmarks, regression detection, structured A/B comparison - as opposed to shipping and hoping. * Strong autonomy, ownership, and technical leadership across teams, including mentoring senior engineers and driving architectural decisions. * Comfortable operating in a dynamic, fast-paced, outcome-driven environment. Great to Have * Hands-on experience with post-training algorithms and infrastructure, including SFT and RL. * Experience with content understanding and/or information retrieval in structured-document-heavy domains. * Prior work on RAG systems involving data sources in different formats (Google Docs, PDFs, DOCX, etc.). ## Description Ironclad's Intelligence Platform team owns Agent Assistant, Conversational Search, and Content Understanding - the systems that help customers and AI agents understand, find, and act on the right contract information. These are the flagship AI capabilities of our product, built and operated by a combined team of ML and ML infrastructure engineers. We have multiple roles open, and are hiring a range of levels - Staff and Senior Staff. As a Staff or Senior Staff Engineer, Agentic Search, you'll own the architecture that combines LLMs and retrieval systems to answer complex, ambiguous questions about a customer's contracts, and you'll set the technical direction that other engineers across the AI organization build on. You'll partner closely with product, applied science, and engineering leaders to raise the company's search quality bar, and you'll bring the technical depth and eval-driven rigor to turn ambiguous problems into shipped, measurable improvements. Scope and ownership will be calibrated to level. What You'll Do * Own agentic search architecture. Design and evolve the systems that combine LLMs and retrieval to produce optimal answers to complex or ambiguous questions. * Drive eval-driven development. Design and run the benchmarks and experiments that measure search quality, and use that feedback to continuously improve the system. * Raise the search quality bar. Contribute to and influence the company's overall search quality standard. * Own content understanding and ingestion. Turn raw documents into processed data that retrieval systems can consume, by building and using NLP/LLM models and pipelines. * Set technical direction. Define architectural decisions and technical direction that other engineers across the AI organization build on. ## Related Videos - [Small, Secure, Interconnected: The next Internet Protocol](https://www.wearedevelopers.com/videos/100062-small-secure-interconnected-the-next-internet-protocol) - [Add Location-based Searching to Site with ElasticSearch](https://www.wearedevelopers.com/videos/77-add-location-based-searching-to-site-with-elasticsearch) - [Carl Lapierre - Exploring Advanced Patterns in Retrieval-Augmented Generation](https://www.wearedevelopers.com/videos/1235-carl-lapierre-exploring-advanced-patterns-in-retrieval-augmented-generation) - [GitHub Next and the Future of Coding - Idan Gazit](https://www.wearedevelopers.com/videos/1903-github-next-and-the-future-of-coding-idan-gazit) - [Building an AI-Ready Content Lake: Scaling RAG and Document AI Beyond Demos](https://www.wearedevelopers.com/videos/1977-building-an-ai-ready-content-lake-scaling-rag-and-document-ai-beyond-demos) - [Serverless Observability: where SLOs meet transforms](https://www.wearedevelopers.com/videos/854-serverless-observability-where-slos-meet-transforms) ## Related Articles - [Dev Digest 120 - Apple and peers](https://www.wearedevelopers.com/magazine/455-dev-digest-120-apple-and-peers) - [Dev Digest 137 - AI'm not sure about this](https://www.wearedevelopers.com/magazine/485-dev-digest-137-ai-m-not-sure-about-this) - [How We Built a Worry-Free System That Runs for 10+ Years – And What We’d Do Again](https://www.wearedevelopers.com/magazine/751-how-we-built-a-worry-free-system-that-runs-for-10-years-and-what-we-d-do-again) - [Dev Digest 121 - AI goes offline](https://www.wearedevelopers.com/magazine/456-dev-digest-121-ai-goes-offline) - [Dev Digest 132 - Binging WADFlix?](https://www.wearedevelopers.com/magazine/473-dev-digest-132-binging-wadflix) - [What Are Large Language Models?](https://www.wearedevelopers.com/magazine/304-what-are-large-language-models)