Sr. AI Engineer

Ovative Group•Minneapolis, MN
•$90,000 - $132,000•Hybrid

About The Position

Ovative Group is hiring a Sr. AI Engineer for its AI OS & Agentic Execution team. The team builds agents that perform operational work, plan multistep tasks, use tools across company systems, check their output, and involve people when judgment or approval is needed. Agents run on a shared, governed foundation designed for reuse. The engineer will design, build, and operate production agents and agentic workflows that automate processes currently involving manual movement of information between systems. The role involves choosing between autonomous agents, structured agentic workflows, and conventional code; designing agent tools, context, memory, orchestration, and permissions; and demonstrating reliability through evaluation. The engineer will also help build shared services and infrastructure for agents, and support deployed systems in production. The role works day to day with the Lead AI Engineer and AI Product Owner, and partners with Platform Engineering, Information Security, and the media operators who use the systems. The position is based around Ovative’s Minneapolis, Chicago, or New York offices, with set in-office days each week and remote work on other days.

Requirements

  • At least 4 years of software engineering experience, with backend or platform work as a core part of that experience.
  • Personally designed, built, and operated multiple autonomous agents in production that took actions through tools or APIs, and can explain their loops, tools, failures, and subsequent improvements.
  • Built agentic workflows combining model steps with deterministic code, and can explain the choice of workflow pattern and why it was used.
  • Hands-on tool-calling and agent-orchestration experience through MCP, an agent SDK such as the Claude Agent SDK or OpenAI Agents SDK, or a framework such as LangGraph; able to explain when a framework is useful and when plain code is preferable.
  • Experience designing tools for agents and engineering context, including prompt contents, retrieval, persistent memory, and methods to prevent long-running tasks from drifting.
  • Experience with multi-agent systems or subagent delegation, including work handoffs and debugging.
  • Experience managing state in long-running work using queues, workflow engines, checkpoints, or human approval steps that pause and resume.
  • Measured agent quality using evaluation suites, including task success and tool use, rather than spot checks.
  • Instrumented agents with tracing and logging and used traces to debug a real failure.
  • Strong Python skills and comfort with SQL and Postgres.
  • Experience building backend services with authentication, authorization, and least-privilege access.
  • Experience integrating with third-party APIs that write data, including handling retries, idempotency, and rate limits.
  • Experience with at least one major cloud platform (AWS, GCP, or Azure), containers, and CI/CD.
  • Ability to explain technical tradeoffs clearly to operators and engineers.
  • Applicants must be legally authorized to work in the United States for any employer on a full-time basis, without current or future need for visa sponsorship.

Nice To Haves

  • Experience with the Claude API, Claude Agent SDK, or Claude Enterprise.
  • Experience building or operating an AI or MCP gateway, policy enforcement for agents, or agent registries.
  • Databricks experience, including Lakebase or Unity Catalog.
  • Experience with advertising platform APIs such as the Meta Marketing API or Google Campaign Manager 360, or with ClickUp and similar work-management APIs.
  • Experience operating AI systems in governed or compliance-sensitive environments, including SOC 2.
  • Background in marketing, media, or professional services.

Responsibilities

  • Design and build autonomous, tool-using agents that plan multistep work, select and call tools, observe results, adapt, and finish or escalate to a person.
  • Design agent loops, including planning, task decomposition, stopping conditions, error recovery, and human escalation.
  • Design reliable agent tools with clear names, constrained inputs and outputs, useful error messages, and appropriate granularity.
  • Engineer agent context and memory, including step-level context, cross-step and cross-session retention, and retrieval of business and client knowledge.
  • Orchestrate multiple agents, including delegating work to specialist subagents and using reviewer agents to check other agents’ work.
  • Convert complex business processes into agentic workflows using patterns such as chaining, routing, parallel steps, orchestrator-and-worker, and evaluator-and-optimizer loops.
  • Choose where models provide judgment and where deterministic code should control the process, keeping exactness-critical steps out of model control.
  • Build human-in-the-loop checkpoints for review, approval, correction, or rejection at appropriate points.
  • Handle third-party system integrations, including rate limits, retries, idempotent writes, partial failures, and safe rollback.
  • Build and operate shared agent services, including gateway action checks, identity and client-level access, durable pause-and-resume state, audit logging, and a kill switch.
  • Expose approved tools through MCP servers with scoped credentials, and help establish the agent registry and observability layer.
  • Build evaluation suites for task success, tool selection, trajectory quality, and regressions, and run them in a sandbox before live client accounts are affected.
  • Instrument agents with traces and metrics that show actions, reasoning context, duration, and cost.
  • Support production agents and turn failures into evaluation cases or guardrails.
  • Work with operators to map actual processes and design agents around their needs.
  • Partner with Information Security on reviews, MCP policies, and logging standards, and with Platform Engineering on environments, Databricks, and Lakebase.
  • Document agent designs and decisions so future engineers and agents can build on the work.

Benefits

  • Annual bonus opportunity of up to 20% of base salary.
  • Competitive health insurance options, including coverage for same-sex partners.
  • Family formation benefits, including reimbursement options for fertility, pregnancy, and parenting needs.
  • Generous paid vacation policy.
  • 401(k) match program.
  • Monthly stipend for mobile phone and data plan.
  • Sabbatical program.
  • Charitable giving program with time and financial match.
  • Access to office spaces in Minneapolis, New York City, and Chicago.
  • Frequent paid travel to the Minneapolis headquarters for company and team events and in-person collaboration.
  • Shenanigan’s Day.
© 2026 Teal Labs, Inc
Privacy PolicyTerms of Service