Senior AI Engineer, MapGPT
New
J
JobgetherAI, Geospatial Technology
CanadaFull-TimeSenior
Salary not disclosed
Apply NowOpens the employer's application page
Job Details
- Experience
- At least 5 years of software engineering experience; at least 2 years of experience shipping LLM-powered features.
- Required Skills
- PythonSQLTypeScriptData engineeringLLMDistributed Systems
Requirements
- Bachelor's degree in a STEM discipline.
- At least 5 years of software engineering experience, including production ownership of services, pipelines, or SDKs.
- At least 2 years of experience shipping LLM-powered features to real users in production with reliability requirements and on-call responsibilities.
- Strong data engineering experience, including SQL and at least one distributed processing framework.
- Experience building and operating data pipelines where data accuracy is critical.
- Strong understanding of tool calling and agent orchestration, including common failure modes.
- Advanced proficiency in Python or TypeScript.
- Direct experience or deep knowledge of designing evaluations for non-deterministic AI systems.
- Familiarity with multiple agent or AI application harnesses.
- Experience diagnosing latency across distributed request paths and optimizing for performance.
- Strong technical judgment and comfort working through ambiguous problems.
- Experience collaborating across engineering and product teams.
Responsibilities
- Own the technical design and delivery of multi-component AI systems, taking accountability for the quality and reliability of production releases.
- Define product behavior, establish measurement frameworks, and develop evaluation systems for non-deterministic AI behavior.
- Create datasets and evaluation cases using real-world usage data and regression results to determine release readiness.
- Build and maintain data pipelines covering ingestion, conflation, entity resolution, and streaming processing.
- Design feedback loops that transform product usage and system failures into actionable data and new evaluation cases.
- Optimize systems against latency and cost targets through techniques like streaming, caching, and model routing.
- Build and improve model harnesses and agent orchestration systems, accounting for failure modes like hallucination and partial success.
- Participate in on-call rotation and contribute to code and technical design reviews.
View Full Description & ApplyYou'll be redirected to the employer's site
