Staff Software Engineer

Pelagon is a European defence technology company building Autonomous Maritime Interdiction systems. It brings together sensing, AI, software and maritime engineering to help protect critical infrastructure and maritime environments. The role is to own the software running in Pelagon's operations centre, including data infrastructure, processing services, internal tools and customer-facing applications. The engineer will architect the overall platform and build it with the team, from data ingestion and storage to machine-learning workloads, operational applications and deployment, acting as the technical interface between embedded systems and AI teams. Software must run on infrastructure operated by Pelagon or its customers, including self-hosted and on-premises environments, with robustness, operability and cybersecurity designed in from the outset.

Responsibilities

Software architecture and delivery: define the architecture of core services, APIs, internal tools and customer-facing applications, make and document technical decisions, write production code and take responsibility for the system from design through operation; Data infrastructure at scale: build reliable ingestion, storage and retrieval for large volumes of sensor and operational data, choose and operate databases and storage systems, design data models, and manage data quality, migrations, retention and performance; Processing and machine-learning workflows: build infrastructure to schedule, execute and monitor data-processing and ML jobs, handle dependencies, resource allocation, retries, reproducibility and failure recovery, working with the AI team to bring workloads into production; Deployment on infrastructure we control: make the platform reproducibly deployable on Pelagon and customer infrastructure, own infrastructure automation, packaging, configuration, release pipelines, upgrades and rollback without depending on managed cloud services; Reliability and operations: establish observability, performance testing, capacity planning, backup and recovery, diagnose production issues and build systems that recover from interrupted connectivity, partial failures and faulty inputs; Cybersecurity throughout the lifecycle: build security into architecture, implementation, deployment and maintenance, apply threat modelling, least-privilege access, authentication and authorisation, network isolation, secrets management, encryption and audit logging, establish secure development practices, vulnerability management, patching and incident readiness; Integration across teams: define clear interfaces and data contracts with embedded systems and AI teams, own integration and end-to-end testing; Engineering effectiveness: set a high standard for code, testing, documentation and technical judgement, mentor engineers and use AI-assisted development tools while retaining rigorous review, security checks and control over sensitive code and data

Requirements

Extensive experience building and operating robust, data-intensive production software, with architectural ownership across multiple components and teams; Strong hands-on software development skills, with the ability to move from system design to implementation and difficult production debugging; Deep experience with data ingestion, databases, storage systems and distributed processing, including the practical challenges of performance, consistency and scale; A strong DevOps and systems background: Linux, networking, containers, orchestration, infrastructure as code and automated delivery; Proven experience deploying and maintaining software on self-managed infrastructure, private infrastructure or customer premises, including operating the services that managed cloud platforms would otherwise provide; Experience building or operating job orchestration and data-processing platforms, and integrating ML workloads into production systems; Practical cybersecurity expertise and a record of applying it to production software and infrastructure, including identifying threats, making security trade-offs explicit and implementing effective controls; The judgement to choose maintainable solutions, anticipate failure modes and balance delivery speed with long-term reliability; The ability to lead technically across disciplines, clarify ambiguous requirements and take ownership of the overall outcome; Effective use of modern AI engineering tools, with the discipline to verify generated code and protect confidential information

Nice to have

Sensor, time-series, streaming or geospatial data; Compute-intensive workloads, GPU infrastructure or ML platform engineering; Enterprise software deployed across multiple customer environments; Systems with demanding availability, security or connectivity constraints; Operator-facing applications and software connected to physical systems