Agentic Software Engineer: Production AI & Automation

Uber; United States; On-site; USD 171,000 - 190,000; Full time

Job summary

Uber’s Agentic Pods is seeking an Engineer to design, build, and operate production‑grade agentic systems that improve complex operational workflows. You will embed with operators, own solutions from discovery through production rollout, and work across backend, reliability, and safety disciplines.

You will partner with security, privacy, legal, and data teams to ensure safe, explainable, and recoverable agentic actions, while building reusable platform components and scalable patterns for

Qualifications

• Bachelor’s degree or equivalent in computer science or related field, or 1 year of full‑time software engineering experience.
• Strong experience designing and building production backend or distributed systems.
• Experience owning and improving production systems including monitoring and incident response.

Responsibilities

• Embed with operational teams to understand workflows and identify opportunities for agentic AI or automation.
• Design, build, and iterate on production agentic systems and integrations.
• Define reliability targets and implement observability across agent decisions, model behavior, dependencies, and workflow outcomes.
• Build reusable components and scale patterns for cross-team use.

Skills

Programming (C/C++/Java/Python/Go)

Production backend systems

Observability and monitoring

Incident response and debugging

Education

Bachelor’s degree or equivalent in Computer Science, Engineering, Mathematics or related field

Tools

APIs

Data pipelines

Job description

Uber’s Agentic Pods is seeking an Engineer to design, build, and operate production‑grade agentic systems that improve complex operational workflows. You will embed with operators, own solutions from discovery through production rollout, and work across backend, reliability, and safety disciplines.

You will partner with security, privacy, legal, and data teams to ensure safe, explainable, and recoverable agentic actions, while building reusable platform components and scalable patterns for