Staff Software Engineer, Agentic Analytics

Role overview

Responsibilities

• Own architecture for converting business questions into governed queries and verifiable answers, with correct context, approved definitions and joins, user permissions, and evidence.
• Build production services, agent data tools, context retrieval, analytics skills, and answer APIs; meet reliability, latency, cost, security, and observability standards.
• Integrate the system into Slack, North, IDEs, and other agents.
• Define and run evaluations, including ground-truth and held-out questions, clarification or decline cases, scoring, coverage, error severity, and regression tests.
• Create agent-usable context from metric definitions, approved joins, generated table documentation, glossaries, and examples; convert corrections into reviewed definitions and tests.
• Roll out to marketing, revenue, finance, product, and other domains; track adoption, accuracy, turnaround, and correction effort.
• Package and document tools, skills, context templates, and evaluations for use by other teams.
• Hire and lead engineers, set technical standards, review work, and develop team members.
• Partner with data platform, model evaluation, security, and domain owners on data models, evaluation, access, definitions, and rollout.

Qualifications

• 10+ years building and operating production analytics or software systems, including technical leadership of a team or systems spanning teams.
• Experience shipping production LLM agents, including tool use, MCP or similar, retrieval and context engineering, guardrails, and observability.
• Hands-on experience building natural-language or agentic analytics relied on by real users, with evidence it was correct.
• Depth in LLM evaluation, including golden sets, result-level comparison, abstention metrics, error taxonomies, and regression testing in CI.
• Strong Python and SQL; experience designing and operating production services and APIs on a major cloud.
• Experience with cloud warehouses, dbt, and semantic layers; Google Cloud is a plus.
• Experience implementing identity-aware agent access using the requesting user's permissions and maintaining an audit trail.
• Experience hiring and developing engineers; ability to operate in ambiguity, lead through partners, and write decision-making design documents and evaluation reports.
• Interest in AI research and hands-on technical work.

Compensation

• Compensation range is CAD 250,000–325,000.

Benefits

• Benefits include a weekly lunch stipend, health and dental coverage, mental health budget, retirement matching, and parental leave top-up.
• Perks include six weeks of paid vacation, enrichment and learning benefits, and travel support for remote employees attending other offices or the annual offsite.

Workplace

• Cohere describes itself as remote-friendly and has offices in Toronto, London, New York City, San Francisco, Montreal, Paris, Berlin, and Seoul.
• Employees away from offices can receive a co-working benefit; all employees receive a CAD 500 home-office stipend.