Computer Vision Flight Software Lead

About the Company
Established in 2017, Infinite Orbits is a NewSpace company, based in Toulouse, pushing the boundaries of in-orbit services with cutting-edge spacecraft driven by its own Rendez-Vous solution, an advanced Autonomous Vision-Based Navigation Solution. IO has two generations of missions
• "Orbit Guard" provides Inspection and Space Situational Awareness services in GEO.
• "Endurance" is a docking service capable of extending the life of GEO telecom satellites, by taking over AOCS takeover for 5 years.

About the Role
Infinite Orbits is hiring a hands-on Flight Software Lead for its Computer Vision & AI team in Toulouse, to take our C/C++ vision stack from flight-software-compatible to flight-ready and build the process that gets it there fast and safely. You own the path from research code to code that flies. Our CV team builds vision-based navigation for satellite rendezvous, proximity operations and autonomous docking: GEO life-extension docking on a Launch Adapter Ring, and LEO microsatellite RPO. The algorithms are strong. What is missing is the person who turns them into flight software. You will define what "flight-grade" means for AI-driven computer vision at Infinite Orbits, set the quality bar, automate the checks, and coach every engineer until the team writes to that bar by default. We are a new-space company. We want rigour where it protects the mission and modern software practice everywhere else: automated pipelines, fast feedback, clear gates. We do not want a 1990s waterfall process. Your objective: take our C and C++ vision code from flight-software-compatible to flight-ready, and build with the CV Lead a repeatable process that makes shipping it fast and safe. Alternate titles: Flight Software Integrator (CV), Flight Coder, CV Flight Software Lead. Where you sit : You report to the Head of Computer Vision & AI and and you are the liaison to the FSW team, working alongside a mid-level CV and embedded engineer and an FPGA engineer. You are the single point through which CV code reaches the spacecraft. Team What you do with them : CV Algorithms Understand every algorithm and the codebase deeply; review, refactor and harden code for flight. AI and V&V Agree how trained models and test campaigns become verified, shippable builds. Simulators (SRS, UIS) Use simulation and hardware-in-the-loop data as release evidence. Flight Software team Define the interface to the platform and other subsystems; deliver the final CV flight code into their build. FPGA team Agree the split between PolarFire SoC CPU and FPGA fabric; integrate accelerated blocks. GNC and Docking Agree CV output formats, rates, latencies and health flags consumed downstream. Camera equipment Understand sensor capabilities and limits as they affect the pipeline.

Job Description
What you will do : Define flight-grade for us
• Work out, with the CV Lead, what "flight-grade" and "ready to fly" mean for AI-based vision code at Infinite Orbits, and write it down as requirements.
• Set the quality levels and release gates code must pass before it leaves the team.
• Decide where strict standards apply and where new-space pragmatism is acceptable, and enforce the result. Build the pipeline that vets every release
• Design the automated CI/CD tests: MISRA C/C++ compliance, safe numerics, static analysis, memory safety, coverage and timing checks.
• Make sure no code leaves the team without passing that pipeline.
• Build an automated delivery pipeline for the CV stack so stable, versioned builds ship to the Flight Software team. Own the flight architecture
• Choose software architectures and memory-management strategies for deterministic, resource-constrained targets.
• Design over-the-air updates for the CV stack: packaging, versioning, validation and rollback.
• Design internal CV FDIR (fault detection, isolation and recovery) and the health data it reports to the platform. Bridge to the platform and hardware
• Act as the CV team's liaison to Flight Software: define how CV interfaces with the platform and other subsystems.
• Understand the limits of the PolarFire SoC, the FPGA fabric, the flight software and the cameras, and design within them. Code, review and teach
• Write code yourself, including hand-written code for edge devices.
• Dive deep into other engineers' code and understand it at implementation level.
• Run one-on-ones, in-person coaching and team presentations on how to write code to our standard.
• Build a step-by-step path that brings the whole team to flight-grade code. Requirements Must have
• 8+ years of professional software engineering. We weigh what you have built and shipped more than the count of years.
• A proven track record of deploying C and C++ on edge hardware in autonomous systems (spacecraft, drones, autonomous vehicles, robotics), more than once. You have taken code from development to running unattended in the field, and handled memory, timing and numerical behaviour on the target.
• Space heritage: you have written code that flew. You know why the flight standards exist, without being tied to legacy, heavyweight ways of applying them.
• Flight, safety-critical or high-reliability software practice: coding standards such as MISRA C/C++, and a qualification framework such as ECSS-E-ST-40 / ECSS-Q-ST-80, DO-178C or ISO 26262.
• Built CI/CD for embedded code: static analysis, cross-compilation, unit and integration tests, coverage, hardware-in-the-loop.
• Understanding of SoC and FPGA architectures: CPU/fabric partitioning, caches, DMA, interrupts, fixed-point versus floating-point trade-offs.
• Ability to read and reason about computer vision and numerical code in depth. You do not need to invent the algorithms.
• Experience mentoring engineers through code review, pairing and teaching, and the patience to do it every week.
• Fluent English. French is a plus. Nice to have
• Experience at an advanced software company, bringing modern engineering practice: trunk-based development, automated testing, fast release cycles.
• Microchip PolarFire SoC (RISC-V) experience, or similar (Xilinx Zynq, Versal).
• Neural-network inference on edge hardware: quantisation, ONNX, FPGA acceleration.
• Experience designing FDIR or over-the-air update systems.
• Embedded Linux or RTOS work (FreeRTOS, RTEMS, Zephyr).

What we offer
• The opportunity to be part of an international team transforming the space industry.
• A creative and innovative work environment where ideas turn into reality.
• Competitive salary and benefits.