Software Engineer

Hyades; Auckland; Hybrid; NZD 120,000 - 180,000; Full time

Job summary

Hyades is seeking a Software Engineer to build the BlazerOps platform from the ground up, owning hard infrastructure challenges and delivering a durable product for ML teams. You will work across the full stack, from Python SDKs to orchestration layers, to web interfaces, ensuring reliability and a great user experience.

You will collaborate with the research team to translate experiments into production features and contribute to documentation.

Qualifications

• Solid experience with Python and at least one compiled language (Go, Rust, or C++).
• Strong understanding of distributed systems, TFQ: task queues, state management, fault tolerance in production environments.
• Experience building and operating services on cloud infrastructure (AWS or GCP).
• Familiarity with ML workflow tooling: experiment tracking, data versioning, or pipeline orchestration.

Responsibilities

• Build and maintain the core BlazerOps platform: pipeline orchestration, experiment tracking, and storage layer.
• Design APIs and data models that are clear and easily extended as the platform grows.
• Architect and implement services that coordinate automated experimentation across distributed compute environments.
• Ensure platform reliability, observability, and performance at scale.
• Maintain and extend the Python SDK that connects ML practitioners to the BlazerOps platform.
• Build CLI tooling for complex platform operations.
• Collaborate with research to translate capabilities into production features.
• Contribute to technical documentation for design partners.

Skills

Python

Go

Rust

C++

Tools

AWS

GCP

Job description

We are looking for a Software Engineer to help build the BlazerOps platform from the ground up. Someone who takes ownership of hard infrastructure problems, writes code that lasts, and cares deeply about the experience of the ML teams who will use what you build.

About Hyades

Hyades is a Spatial AI research lab on a mission to make machines understand the physical world. We are building the foundational intelligence layer for spatial data and machine learning, and the MLOps platform that delivers that intelligence to enterprise geospatial ML teams.

BlazerOps is a platform that handles everything from raw data ingestion to model deployment, with intelligence built in at every stage. You will be working across the full stack: from the Python SDK that data scientists use every day, to the orchestration layer that runs experiments at scale, to the web interfaces that make the platform accessible.

What you will do

Platform and Infrastructure

• Build and maintain the core BlazerOps platform: pipeline orchestration, experiment tracking, and the Living Representation storage layer.
• Design APIs and data models that are clear, consistent, and easy to extend as the platform grows.
• Architect and implement the services that coordinate automated experimentation across distributed compute environments.
• Ensure platform reliability, observability, and performance at the scale our design partners operate.
• Maintain and extend the Python SDK that connects ML practitioners to the BlazerOps platform.
• Build CLI tooling that makes complex platform operations feel simple and composable.
• Collaborate with the research team to translate experimental capabilities into stable, production-grade platform features.
• Contribute to technical documentation that helps design partners get the most from the platform.

Requirements

• Solid experience with Python and at least one compiled language (Go, Rust, or C++).
• Strong understanding of distributed systems: task queues, state management, and fault tolerance in production environments.
• Experience building and operating services on cloud infrastructure, preferably AWS or GCP.
• Familiarity with ML workflow tooling: experiment tracking, data versioning, or pipeline orchestration.
• Comfortable owning a system end-to-end: from initial design through deployment and on-call support.

Nice to have

• Experience with geospatial data formats: STAC, COG, GeoTIFF, or similar.
• Familiarity with ML compute frameworks: Ray, Dask, or distributed PyTorch training.
• Experience building developer-facing SDKs or CLI tools.
• Background in data engineering or large-scale data processing pipelines.
• Contributions to open-source projects in the ML or geospatial ecosystem.
• Experience integrating with orchestration platforms: Dagster, Prefect, or Airflow.