[Hiring] Software Engineer @Prelude

Role Description

Origin is hiring software engineers to help us build across our stack: the agent that collects the data, the service that makes sense of it, and the interface that presents it to customers.

• You will own design and implementation from idea through release.
• The role demands a high degree of agency, product taste, and raw technical aptitude.
• You will work across all of it, from high-performance data ingestion through inference pipelines to the API design and customer-facing surfaces on top.
• You will be a heavy user of the product you are building, and the friction you feel in your own workflow becomes the next thing we ship for customers.

This is an AI-native engineering team: most of the code we ship is written by agents. You should already be fluent in the agentic coding tools (Claude Code, Codex, OpenCode) and able to build your own when an existing one does not fit.

Responsibilities

• Obsess over customer problems and ship for velocity and impact over technical pedantry.
• Ship across the stack - agent, backend, infrastructure, and frontend - wherever the work is that week.
• Build internal agents that automate the repeatable parts of engineering and distribute them across the team.
• Define how AI-native engineering works at Origin: the context, permissions, and tooling our agents need to do real work.
• Design and operate distributed systems using AWS, EKS, Clickhouse, and Kafka against real production SLAs.
• Drive technical decisions on storage schemas, API contracts, and the patterns that hold up at scale in a multi-tenant environment.
• Work alongside research and systems-internals engineers to turn their work into reliable production software.

Qualifications

• AI-native engineering practice: prolific use of Claude Code, Codex, and OpenCode, with the judgment to know when an agent will go off the rails and the skill to redirect it.
• Track record of building data-intensive systems at scale (millions of concurrent operations, high-throughput event processing, distributed coordination).
• Deep understanding of distributed systems principles: streaming, batching, backpressure, recovery, and consistency models.
• Performance engineering mindset: complexity analysis, profiling-driven optimization, and discipline under resource constraints.
• Modern systems programming experience, Rust preferred (we value problem-solving ability over specific language expertise).
• Strong working fluency in Python across services, inference pipelines, and tooling.
• Production experience with cloud infrastructure (AWS preferred), Kubernetes (EKS or equivalent), and event streaming (Kafka or equivalent).
• Comfortable working in our React frontend to deliver functionality to end users.
• High agency, with a track record of identifying important work and finishing it.

Nice to Haves

• Background in real-time monitoring or observability infrastructure (Palantir, Datadog, Honeycomb, Chronosphere, the data side of Stripe or Cloudflare).
• Background in high-throughput systems (high-frequency trading, real-time telemetry, low-latency distributed systems).
• Background in AI infrastructure or applied ML, particularly inference pipelines, embedding and topic systems, or agent orchestration.
• Prior founding-engineer or early-stage startup experience.
• Open-source contributions or published research in AI tooling, agent frameworks, or developer infrastructure.
• Data engineering experience with columnar databases or OLAP / time-series systems.
• Deep OS internals knowledge (Windows ETW, Apple ESF, eBPF).
• Cross-platform systems expertise (Windows and macOS).

Benefits

• Fully remote team across the US & Canada.
• Built on trust, autonomy, and excellence.
• Generous healthcare.
• Flexible PTO.
• Home-office support.