> Markdown version of [/jobs/ext/2046545-machine-learning-operations-architect](https://www.wearedevelopers.com/jobs/ext/2046545-machine-learning-operations-architect). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # Machine Learning Operations Architect - **Company:** Inbrain Neuroelectronics - **Location:** Barcelona, Spain - **Contract:** Permanent contract - **Skills:** Training Data, Clinical Data Repository, Continuous Integration, Python (Programming Language), Management of Software Versions, Fast Healthcare Interoperability Resources, Model Validation, Kubernetes, Information Technology, Performance Monitor, Machine Learning Operations, DO-178B, Software Version Control - **Published:** August 14, 2026 - **Apply:** https://www.jobleads.com/es/job/e461f026689c5195be8230a8ea14ef6f0 ## About the Role * Bachelor's or Master's degree in Computer Science, Mathematics, Physics, Electrical Engineering or a related field. * At least 4-5 years of Hands-on ML Ops engineering experience (preferably in medtech industry): model versioning/registries (MLFlow or equivalent), CI/CD for ML, containerized deployment (Kubernetes), and workflow orchestration (e.g. Argo or equivalent). * Proven experience shipping ML into a regulated or safety-critical environment (medtech - priority, automotive, aerospace), has designed systems to automatize design-control processes (e.g. IEC 62304, ISO 13485, DO-178C, ISO 26262) and has used them from the inside. * Strong infrastructure-as-code skills and comfort directly supporting data scientists to productionize research code written in Python. * Fluency in English required (English is company language). * Direct experience with continuous/adaptive learning systems under a change-control or PCCP-like framework, delivering personalized evolving models in contrast to single model validation for universal use releases. * Comfortable operating at the intersection of engineering and regulatory/clinical teams, translating design-control requirements into concrete technical implementation. * Pragmatic about scope and sequencing - able to prioritize the governance/compliance work. * Strong written documentation skills, given the role's heavy emphasis on producing auditable change-control and validation records., * Experience with FDA's Predetermined Change Control Plan (PCCP) guidance or equivalent adaptive-SaMD regulatory frameworks (e.g. through an imaging/diagnostic AI company that has filed one). * Background in adaptive neurostimulation, closed-loop deep brain stimulation or closed-loop diabetes management with insulin pumps. * Familiarity with clinical data standards (FHIR, SNOMED CT, NWB/BIDS) and ISO 14155/MDR/ICH E9 evidence frameworks. We are looking for someone who Is ready to proactively bring new ideas to the team, push boundaries, and constantly look for innovation. At INBRAIN we believe in shared success and diverse ways of thinking, here you'll learn, grow, and advance in an innovative culture ## Description * Design and build the model versioning, lineage, and validation-evidence system that ties every trained model to its training data, the data source origin, risk management, and clinical evidence. * Draft and maintain a Predetermined Change Control Plan (PCCP)-style framework defining permitted modification types, bounds, and automated re-validation protocols for adaptive models, in collaboration with clinical, regulatory, product and software stakeholders. * Design and configure the CI/CD pipeline for training, validating, and promoting models, running on infrastructure provided by the Software team, with promotion gates across Development, Testing, Acceptance and Production (DTAP) tiers. * Build drift and performance monitoring for models and define triggers for scheduled vs. drift-triggered retraining as part of post-market surveillance activity. * Deployed and maintained processes and services versioning and tracing them against design-control processes. ## Related Videos - [Effective Machine Learning - Managing Complexity with MLOps](https://www.wearedevelopers.com/videos/185-effective-machine-learning-managing-complexity-with-mlops) - [Fireside Chat: Deep Learning, Deep Impact: Harnessing AI for Language Innovation](https://www.wearedevelopers.com/videos/612-fireside-chat-deep-learning-deep-impact-harnessing-ai-for-language-innovation) - [Understanding Kubernetes in a visual way](https://www.wearedevelopers.com/videos/100085-understanding-kubernetes-in-a-visual-way) - [MLOps - What’s the deal behind it?](https://www.wearedevelopers.com/videos/392-mlops-what-s-the-deal-behind-it) - [Exploring 5 Key Applications of AI Abundance with Blockchain Assurance](https://www.wearedevelopers.com/videos/971-exploring-5-key-applications-of-ai-abundance-with-blockchain-assurance) - [Instant KAI Sandboxes with vCluster: Multi-Tenant, Multi-Scheduler GPU Sharing](https://www.wearedevelopers.com/videos/100333-instant-kai-sandboxes-with-vcluster-multi-tenant-multi-scheduler-gpu-sharing) ## Related Articles - [MLOps – What’s the deal behind it?](https://www.wearedevelopers.com/magazine/125-mlops-what-s-the-deal-behind-it) - [MLops – Deploying, Maintaining And Evolving Machine Learning Models in Production](https://www.wearedevelopers.com/magazine/115-mlops-deploying-maintaining-and-evolving-machine-learning-models-in-production) - [Are Brain-Computer Interfaces Good or Bad?](https://www.wearedevelopers.com/magazine/367-are-brain-computer-interfaces-good-or-bad) - [How to Become an AI Engineer](https://www.wearedevelopers.com/magazine/331-how-to-become-an-ai-engineer) - [MLOps And AI Driven Development](https://www.wearedevelopers.com/magazine/82-mlops-and-ai-driven-development) - [Coffee with Developers - Maria Apazoglou - Making AI understandable for all in production](https://www.wearedevelopers.com/magazine/475-coffee-with-developers-maria-apazoglou-making-ai-understandable-for-all-in-production)