18 days ago
Base Salary
$195k - $255k/yr
Responsibilities
- Design the agentic platform’s orchestration, tool registration and permissioning, durable state and memory, sandboxing, and trace-level observability.
- Build long-running agent workflows with multi-step reasoning, tool use, human-in-the-loop checkpoints, and partial-failure recovery.
- Design retrieval-augmented generation systems with hybrid search over legal corpora and citation grounding.
- Build evaluation infrastructure with golden datasets, regression harnesses, and offline and online evaluation loops.
- Establish model-operations practices for routing, latency and cost budgets, prompt and context versioning, caching, and production reliability.
- Build scalable backend services and APIs consumed by other engineering teams.
- Produce technical design documentation and partner with product, research, security, and legal domain experts.
- Operate systems from initial design through production and strengthen engineering standards.
Requirements
- 5+ years of software-engineering experience with depth in backend, distributed, or AI systems.
- Production-scale experience delivering and operating LLM-based applications beyond prototypes or proofs of concept.
- Experience with agentic architectures, orchestration frameworks, and retrieval-augmented generation systems at scale.
- Strong production-level proficiency in Python.
- Experience evaluating agentic-system accuracy and performance.
- Experience designing services and APIs consumed by other engineering teams.
- Experience with observability tooling such as Prometheus, Grafana, or OpenTelemetry.
- Experience operating in containerized cloud environments.
- Strong system-design and architecture experience, including producing technical design documents.
- Excellent written and verbal communication skills and the ability to work with non-technical domain experts and stakeholders.
- Proficiency using AI tools in software development.
- Preferred: experience with fine-tuning, model adaptation, systematic evaluation across model families, large-scale data pipelines, document-processing systems, multi-tenant SaaS, data isolation, startup or scale-up environments, legal technology, enterprise workflows, or knowledge-management systems.
Benefits
- The role is based in the office 3–4 days per week.
- The position may be eligible for an annual bonus.
- Epiq offers enterprise-wide learning and mobility opportunities, flexibility, and benefits described through its benefits program.
Tech Stack
Categories
