Softwareingenieur:in

Syn2core; München; Vor Ort; EUR 90.000 - 120.000; Vollzeit

Zusammenfassung

Syn2core is building AI Copilot for smart glasses to assist assembly-line workers in real time. This role owns CI/CD, release engineering, and IaC for a fleet of GPU edge machines in industrial settings, reporting to the CTO. It’s a growth-centric, year-one mandate designed to scale with the company.

You’ll work alongside the Lead Software Engineer and CTO, turning complex pipelines into stable, auditable releases across client sites, while prioritizing privacy, security, and reliability.

Qualifikationen

• Must have: CI/CD ownership (self-hosted or on-premise).
• Experience with large-scale build systems (e.g. Bazel, Buck).
• Infrastructure-as-code (e.g. Terraform) on a major cloud provider.
• Production observability: metrics, logging, alerting; backup/restore discipline.

Aufgaben

• Make releases reliably end-to-end with versioned, reproducible deployments across client sites.
• Turn client onboarding into configuration with reusable infrastructure modules.
• Provide fleet visibility with metrics, dashboards, and alerting; ensure secure remote access and disaster recovery.
• Enforce privacy and security by default, including device identity and zero-trust considerations.
• Collaborate to build a replay harness turning recorded factory streams into regression tests.

Kenntnisse

CI/CD ownership

Build systems

Infrastructure-as-code

Production observability

Security by design

Tools

Bazel

Buck

Terraform

Jobbeschreibung

At Syn2core, we're building AI Copilot, an AI assistant that runs on smart glasses to support assembly-line workers in real time. Our system streams and processes data in real time, and delivers audio feedback directly to the worker. We're deploying it on the production lines of the automotive industry across the world— privacy-first by design: everything we persist is anonymized at ingestion.

Our long-term vision: the egocentric data we collect is training fuel for the next generation of robot learning – positioning us at the heart of the race towards autonomous robots for industrial deployment in automotive, aviation, aerospace and beyond.

The role

You will own the path between \"code merged\" and \"running verifiably at a client\": CI/CD, release engineering, infrastructure-as-code, and the fleet of GPU edge machines our system runs on inside factories. We built fast, and parts of that chain carry the technical debt of speed — your mission is to make \"what exactly is running at station 12, and is it the same software that passed the acceptance test?\" answerable in one query.

You won't do it alone: you'll work daily with our Lead Software Engineer and the CTO, design decisions happen in the open, and this is a year-one mandate — not a day-one checklist. You're not expected to have done all of it before; you're expected to grow into owning it, and we'll build it with you.

What you'll do

• Make releases boringly reliable — own CI/CD end-to-end until nobody re-runs a pipeline \"just to be sure\"; every edge deployment a versioned, reproducible release with rollout, rollback, and version inventory across client sites
• Turn client onboarding into configuration — consolidate our infrastructure‑as‑code into reusable modules, so a new client site is stood up in days
• Give the fleet eyes and a safety net — metrics, dashboards, and alerting so a station failure pages us before the client notices; backup and disaster recovery with restores that are tested, not assumed; secure remote access done right
• Make privacy and security enforceable — anonymization that is architecturally impossible to skip; device identity and authentication for our real-time streaming connections, from the devices on the line to the backend
• Build the replay harness with us — together with our Lead Software Engineer, turn recorded factory streams into regression tests for the whole pipeline

What you get

• A whole domain that is genuinely yours — architecture decisions included — with a direct line to the CTO, not a ticket queue
• Your work running on real factory lines at major OEMs within weeks, not quarters
• The rare chance to set the operational standard of a company early, instead of inheriting someone else’s

Where you'll be in 12 months

The fleet pages us before a client notices anything. Releases are versioned and boring. A new client site is a configuration file. Every streaming connection belongs to a known device, and compliance audits pass because the pipeline enforces the rules. That's what we'll build together — and you'll be the person who made it so.

Who you are

• Boring beats clever — you take pride in systems nobody has to think about
• Zero tolerance for standing hacks: you see toil, kill it permanently, and document the grave
• Security and privacy by default, not as a hardening phase
• \"Unversioned means unshipped\" — and you hold that line under deadline pressure, including from the CTO
• You learn unusual stacks fast and enjoy it

Your experience

Must have:

• CI/CD ownership, including self-hosted or on-premise environments
• Serious experience with a large-scale build system (e.g. Bazel, Buck) and the appetite to go deep
• Infrastructure-as-code (e.g. Terraform) on a major cloud provider
• Production observability: metrics, logging, alerting — and the backup/restore discipline to go with it

Ways to stand out:

• Streaming media pipelines (e.g. WebRTC) and their security
• Device identity / zero-trust architectures
• Supply-chain security (SBOMs, dependency automation)
• Release engineering at a hardware or embedded company