
Lead AI Engineer
Ovative Group3 days ago
Minneapolis, MN, USAStaff+
Responsibilities
- Own agents end to end, including problem definition, technical design, milestones, implementation, launch, and ongoing operation.
- Lead architecture decisions involving autonomy, agent loops, orchestration, stopping conditions, escalation paths, deterministic code, tools, context, state, and failure handling.
- Design reusable agent systems and shared foundation components including action checks, access controls, durable state, audit logging, kill switches, agent registries, and observability.
- Define security models for agents acting in client accounts, including least privilege, credential scoping, prompt-injection defenses, unsafe-tool-use protections, and approval rules.
- Define evaluation, tracing, metrics, cost tracking, regression testing, and release standards for agent behavior.
- Own production health, incident response, post-incident reviews, and conversion of failures into evaluation cases, guardrails, or design changes.
- Mentor engineers through pairing, design feedback, code review, reference implementations, templates, and documentation.
- Partner with operators, product owners, Platform Engineering, Information Security, and other teams on workflows, architecture reviews, standards, roadmaps, and technical tradeoffs.
Requirements
- At least 7 years of software engineering experience, with backend or platform work as a core part of that experience.
- At least 2 years building LLM-based systems in production.
- Production experience leading multiple autonomous agents or agentic systems that take actions through tools or APIs.
- Experience owning architecture decisions, shared infrastructure or platform components, evaluation frameworks, observability, deployment, and production incident response for AI systems.
- Deep experience with tool calling and agent orchestration through MCP, an agent SDK such as the Claude Agent SDK or OpenAI Agents SDK, or a framework such as LangGraph.
- Experience with tool design, context engineering, long-running tasks, prompt structure, retrieval, memory, multi-agent systems, handoffs, shared state, and cross-agent debugging.
- Experience securing agents that write to external systems and managing LLM cost and latency at production volume.
- Strong Python, SQL, and Postgres skills, plus deep experience building backend services with authentication, authorization, and least privilege access.
- Experience integrating third-party APIs that write data, including retries, idempotency, rate limits, and safe rollback.
- Experience with at least one major cloud provider—AWS, GCP, or Azure—containers, and deployment paths or environments used by other engineers.
- Track record of mentoring engineers, leading technical decisions across teams, and communicating tradeoffs clearly to technical and nontechnical stakeholders.
- Preferred qualifications include Claude API, Claude Agent SDK, Claude Enterprise, AI or MCP gateway design, agent registries, Databricks, Lakebase, Unity Catalog, advertising platform APIs, ClickUp, governed-environment security or compliance reviews, SOC 2, early-stage AI platform experience, and marketing, media, or professional services experience.
Benefits
- Annualized compensation is stated as $125,000 to $175,000 and includes an annual bonus opportunity of up to 25% of base salary.
- Hybrid work from the Minneapolis, Chicago, or New York office on set days each week, with remote work available on other days.
- Access to Minneapolis, New York, and Chicago office spaces, plus frequent paid travel to Minneapolis for company and team events.
- Generous paid vacation, a 401(k) match, health insurance including same-sex partners, family formation benefits, a monthly mobile phone and data stipend, a sabbatical program, and charitable giving time and matching.
- Additional benefits include paid travel for in-person collaboration and a companywide Shenanigan’s Day.