Software Engineer, ML Systems

At Harmonic, we are building a mathematical reasoning engine that operates with absolute precision. While most AI makes maximum-likelihood guesses, Harmonic's Aristotle uses Lean 4 and reinforcement learning to verify its reasoning and results.

Following our Gold Medal-level performance on the 2025 International Math Olympiad (IMO) and the successful resolution of long-standing open problems, we are proving that AI can master the most rigorous domains of human thought. Backed by some of the world’s most prominent investors, we are intentionally scaling an ...

Visit our company blog to learn more about what we are working on!

We are looking for a pragmatic, Software Engineer to own the productionization of our research pipelines. This is an implementation-heavy role designed for an engineer who can take a nascent research idea and build the robust, scalable machinery required to prove it at scale within our cloud infrastructure.
Pipeline Engineering: Build and manage end-to-end ML pipelines (ETL and automated evaluation) that are the bedrock of our RL research.
Bottleneck Resolution: Identify and refactor inefficient research code. You act as the primary engineer ensuring that a promising idea reaches its full potential through scalable code.
Standardization: Establish best practices for versioning, experiment tracking, and CI/CD for ML models to ensure reliability.
Cloud Infrastructure & Observability: Manage the deployment and scaling of workloads on Kubernetes. Implement the tooling and telemetry that allows the team to understand agent behavior and training health at a glance.

BS in Computer Science, a related technical field, or equivalent industry experience
2+ years of relevant industry experience
Expert-level Python skills and a disciplined approach to software engineering (testing, versioning, and modular design).
Experience building and managing end-to-end ML pipelines in a production or research-intensive environment.
Full-stack ML experience: Comfortable moving from data engineering to model debugging.
Experience refactoring research-grade code into high-quality, scalable production packages.
Proven ability to design and implement complex data-loading and evaluation systems for non-deterministic models.
Experience with workflow orchestration tools (e.g., Kubeflow, Airflow, or Metaflow).
Experience managing large-scale experiments on cloud providers (AWS, GCP, or Azure).
Proven track record collaborating directly with researchers to translate algorithmic requirements into engineering roadmaps.
Hands-on experience with containerization (Docker) and orchestration (Kubernetes).