Senior Applied AI Engineer (Agentic Systems)

ABOUT THE ROLE This is a founding-level senior engineering role at a fast-growing, Series A/B-stage B2B SaaS company in the AI compliance and regulatory technology space. The company automates high-stakes marketing and packaging review workflows for enterprise brands in regulated industries — turning lengthy manual approval cycles into fast, auditable, AI-driven processes. As a Senior Applied AI Engineer (Agentic Systems), you will own the technical core of the platform: a safety-focused, deterministic multi-agent orchestration system purpose-built for enterprise compliance. This is a hands-on, in-office role in San Francisco where correctness and trust matter more than demos. You'll work closely with a small, senior team and have an outsized impact on the product's technical direction. WHAT YOU'LL DO Agentic Reasoning & Orchestration - Design and evolve multi-agent LLM systems that decompose complex compliance review tasks into reliable, auditable steps. - Define agent responsibilities, hand-offs, and termination conditions to minimize reasoning drift and maximize consistency across enterprise workflows. Context, Retrieval & Memory Systems - Architect retrieval pipelines using RAG, structured memory, and graph-based retrieval to deliver the right context (brand guidelines, regulations, historical decisions) to agents at the right time. - Balance recall, precision, and latency across large, evolving knowledge bases. Stateful, Asynchronous Workflows - Own long-running, fault-tolerant workflows using Temporal (or similar), ensuring retries, versioning, and determinism across non-deterministic model calls. - Treat agent orchestration as a distributed systems problem — managing state, failures, and observability end-to-end. Evaluation, Safety & Reliability - Build evaluation frameworks using statistical metrics, gold labels, and automated regression testing to prove and maintain system reliability. - Prioritize correctness and trust, particularly in high-risk legal and compliance scenarios. Asset Understanding Pipeline - Collaborate on image and document preprocessing (OCR, layout analysis, vision-language models) to ensure downstream agents receive structured, machine-readable context. - Focus on practical, production-grade solutions over research-oriented approaches. WHAT WE'RE LOOKING FOR Required - 7+ years of professional software/ML engineering experience, with a clear progression of increasing scope and responsibility. - Demonstrated expertise architecting multi-agent AI systems: defining agent roles, multi-step reasoning flows, tool integration, memory/retrieval architectures, and deterministic, auditable workflows. - Strong background in LLM application development and production deployment of AI systems (not just prototyping). - Experience with RAG pipelines, vector stores, and/or graph-based retrieval systems. - Familiarity with stateful workflow orchestration (e.g., Temporal, Prefect, Airflow) applied to distributed AI workloads. - Experience at a startup — or, alternatively, strong exposure to regulated industries (e.g., consumer packaged goods, healthcare, financial services). - Ability and willingness to work in-office in San Francisco, CA at least 3 days per week. - Must be eligible to work in the United States without visa sponsorship (no sponsorship available). Nice to Have - Experience with vision-language models (VLMs), OCR pipelines, or document layout analysis. - Background in compliance, regulatory technology, or marketing/packaging review workflows. - Familiarity with evaluation frameworks for LLM-powered systems (e.g., RAGAS, custom eval harnesses). - Prior experience in a founding engineer or technical lead capacity. COMPENSATION & BENEFITS - Salary: $185,000 – $210,000 per year, depending on experience. - Equity participation in a well-funded, early-stage company at an inflection point. - Opportunity to define the technical architecture of a category-creating product alongside a senior, high-density team. LOCATION - San Francisco, CA — hybrid (in-office 3 days/week, with remote flexibility the remaining days). - Visa sponsorship is not available; candidates must be authorized to work in the United States.