Software Engineer - Applied AI

Software Engineer - Applied AI

We are building the next generation of AI-driven products, and our AI Engineers are rapidly prototyping new agents, workflows, and evaluation tools. We're looking for a Software Engineer who will turn these prototypes into reliable, well-instrumented services and applications that the broader organization can depend on.

This role sits at the intersection of rapid prototying and full-stack engineering, and you will own the architecture, observability, reliability, deployment, and documentation of new services. Your work will enable the team to move fast experimentally while maintaining strong engineering foundations and operational excellence, identifying and addressing any workflow bottlenecks along the way. Clear and consistent communication is key as priorities may change quickly.

Responsibilities

• Collaboration: Work closely with AI Engineers to turn experimental notebooks, scripts, and workflows into reliable tools and services; co-design experiment-friendly systems (feature flags, prompts, model switches, eval hooks) that enable fast but safe iteration.
• Architecture: Own the architecture of tooling and services, defining reusable templates, libraries, and patterns that balance rapid prototyping with maintainability and consistency across the team.
• Observability: Lead observability for AI applications and pipelines-logging, metrics, tracing, alerting, and dashboards-so the team can quickly answer "what is happening right now?" in both experiments and production tools.
• Reliability: Drive reliability and resilience practices for AI systems, including testing strategies, safe failure modes, rollout/rollback approaches, and standards for robust APIs that wrap AI/LLM functionality.
• Infrastructure: Own cloud infrastructure for research tooling (e.g., AWS/GCP), including databases, containerization, CI/CD, and infrastructure-as-code, while setting and upholding engineering standards for production-grade systems.
  Productionize an LLM-based research agent into a monitored microservice with robust APIs, structured logging, evaluation hooks, and end-to-end traces that are appropriately stored for rapid analysis.
• Knowledge Sharing: Document services and systems concisely and effectively, demo tools and code to the team, and create internal Agent tools/skills/playbooks the team can use to speed up development.

Representative Projects

• Productionize an LLM-based research agent into a monitored microservice with robust APIs, structured logging, evaluation hooks, and end-to-end traces that are appropriately stored for fast analysis.
• Stand up a standardized Dagster or Prefect-based orchestration template for AI pipelines with builtin observability, retries, and alerting.
• Create dashboards that surface system health, latency, error rates, and experiment metrics across all AI-backed research tools.
• Implement LLM batch inference pipelines using commercial model endpoints - Develop reusable service and UI templates that other AI Engineers can use to quickly turn successful prototypes into production-ready tools.

Qualifications

• 5+ years of professional software engineering experience with a strong backend or full-stack focus.
• Experience integrating LLMs or other AI/ML systems into applications.
• Deep experience building and operating production services end-to-end (design, implementation, deployment, monitoring, and incident response).
• Strong proficiency with Python and modern service development (e.g., REST APIs, microservices).
• Hands-on experience with observability stacks (logging, metrics, tracing, alerting) and debugging distributed systems in production.
• Experience with workflow/orchestration tools (e.g., Airflow, Dagster, Prefect) and building reliable data or experiment pipelines.
• Cloud deployment expertise (e.g., AWS/GCP), including containers, CI/CD, and infrastructure-as-code.
• Comfortable working in ambiguous, research-oriented environments and translating loosely defined experimental code into maintainable, well-structured systems.
• Strong communication and collaboration skills; able to coach prototype-focused engineers on production best practices and clearly explain tradeoffs to non-infrastructure stakeholders.