2 days ago
Responsibilities
- Design and build production agentic systems and workflows that solve defined problems end to end.
- Own prompts, tools, retrieval, guardrails, observability, cost and latency budgets, evaluation, and rollout.
- Identify operational workflows suitable for AI automation and partner with subject-matter experts to define success criteria and validate outputs.
- Design AI features for compliance, safety, privacy, bias mitigation, human oversight, and content filtering.
- Ship features with documentation, runbooks, evaluation harnesses, and dashboards.
- Build manageable AI ecosystems that help cross-functional teams create and improve workflows and automations.
Requirements
- 7+ years of professional engineering experience, including at least 1 year of hands-on experience building and operating production agentic systems.
- Strong Python familiarity plus one typed language such as TypeScript or Go, with AWS or GCP and containerized deployment experience.
- Several years of production software engineering experience and a track record of leading systems to launch.
- Multiple shipped LLM-powered production features, including experience diagnosing failures and improving systems.
- Practical knowledge of agentic patterns, retrieval and grounding, prompt engineering, structured outputs, and disciplined evaluation.
- Experience standing up and testing local models and closed local-model solutions.
- Strong written and verbal communication skills, including the ability to explain architecture to executives and engineers.
- Comfort using AI coding tools heavily while owning problem framing, design decisions, and verification.
- Advanced retrieval, safety and privacy in user-facing AI, regulated-domain experience, prompt-optimization frameworks, public AI work, or open-source AI tooling experience is preferred.
Categories
About Level
Level builds a learning technology platform that combines established curriculum principles with interactive, game-style design to help students practice academic and life skills. The product runs on a forked, optimized Godot engine engineered to perform reliably on mobile and other constrained hardware. It targets students and educators seeking repeatable, meaningful practice in a format that feels like a game.