Paddy Hobson

Location: Munich (Hybrid/On-site depending on team setup) Type: Full-time Company Overview We are working with a fast-growing Vision / AI company building production software for the food and retail industry. Their systems help customers reduce food waste and improve operational efficiency - supporting sustainability goals through real-time computer vision and automation. With teams across Europe, the US, and Asia, we combine startup pace with real-world deployments at enterprise customers.The Role We are hiring a hands-on engineer to support the delivery of our computer vision / ML products into production. This role sits at the intersection of software engineering applied machine learning, with a strong focus on making ML models run fast, reliably, and at scale on edge devices. You will be responsible for our core video processing framework and deployment stack, working closely with senior ML engineers to ensure model inference performance, stability, monitoring, and field success. While you won’t be expected to design new ML algorithms or lead model training, you will be involved in diagnosing model issues in the field and improving real-world performance through optimization and iteration.This is a great fit for someone who enjoys real-world ML delivery: video streams, edge devices, inference performance, and production debugging.Key Responsibilities ML Model Runtime & Edge PerformanceMake ML models run efficiently on edge devices (latency, throughput, CPU/GPU utilization, memory constraints)Support inference optimization and troubleshooting (profiling, batching, pipeline tuning, runtime constraints)Investigate real-world model failures (data quality, camera placement, lighting, drift, edge-case behaviour) and work with ML engineers on mitigation strategiesEnsure robust model rollout processes: versioning, validation, safe deployment cyclesVideo Pipeline Engineering (Core Focus)Design and optimize real-time video processing pipelines using GStreamerIntegrate and manage streams from IP cameras (RTSP/ONVIF) and USB camerasDebug complex video stream issues (encoding/decoding, dropped frames, jitter, latency, network instability)Deployment & Production OperationsPackage and deploy services using Docker/Podman on Linux-based edge systemsTroubleshoot issues directly on production/staging Linux hosts (logs, profiling, system-level debugging)Implement and maintain monitoring and device health checks (e.g., Checkmk or similar)Event Streaming & InterfacesBuild interfaces between edge devices and online tools / connected machinesWork with event streaming systems (Kafka or similar) for detections, events, and telemetrydeep Kafka expertise isn’t required, but strong conceptual understanding isMust-Have Skills2–5 years of professional experience in software engineering / applied ML engineeringStrong Python skills (asyncio, threading, multiprocessing)Strong Linux skills: CLI, systemd, bash scripting, networking fundamentalsSolid experience with containerization (Docker or Podman)Comfortable debugging real systems remotely and working end-to-end (not just coding isolated modules)Interest in ML delivery and computer vision systems in productionNice to HaveExperience with GStreamer (big plus)Familiarity with computer vision pipelines (OpenCV, image processing)Experience with FFmpeg, RTSP, H.264/H.265, ONVIFWebRTC exposure (low-latency streaming)Kafka / message broker familiarityGerman language skills (corporate language is English)Why This Role is InterestingYou’ll work at the “real ML” layer: getting models running in production environments where conditions are messyStrong collaboration with senior ML engineers, with room to grow into more ML responsibility over timeDirect ownership of the edge inference video stack powering real customer deploymentsInternational team, low bureaucracy, hands-on culture

Location: Munich (Hybrid) Type: Full-time Company Overview We are working with a fast-growing Vision / AI software company building production-grade computer vision systems for the food and retail sector. Their products help customers reduce food waste, improve operational efficiency, and contribute to sustainability goals by enabling better decision-making through real-time visual intelligence. With an international footprint across Europe, the US, and Asia, they combine startup speed with real-world deployments at large enterprise customers. The Role We are looking for a Senior Machine Learning Engineer to take hands-on technical ownership of a key vision product that is moving into field testing with major retail partners in Germany. This role is ideal for someone who enjoys being deeply involved across the entire ML lifecycle - from model development and training through to deployment on edge devices at customer sites. You will act as a hands-on technical lead for the product, driving model improvements, performance validation, and production rollouts. Key Responsibilities Model DevelopmentDesign, implement, and iterate on deep learning architectures for real-time object tracking and event detectionTrain and optimize object detection models using production datasets and domain-specific video dataContinuously improve model robustness for real-world conditions (lighting changes, occlusions, camera angles, motion blur, etc.)Performance Evaluation & ValidationBuild and execute evaluation workflows for accuracy latency benchmarkingTest models using benchmark video datasets and dedicated hardware setupsMonitor model performance regressions and validate incremental updates before releaseDeployment & Integration (Edge / Production)Own the technical process of deploying model updates into production systemsEnsure stable integration of models into the wider software stack running on-siteSupport field testing cycles, troubleshooting and optimizing performance on edge devicesTooling & PipelinesMaintain and improve internal pipelines for:automated model trainingdata versioningperformance testingreproducible experimentationDrive best practices across model development and deployment workflowsRequirements5–8 years experience in Machine Learning / Deep Learning / Computer VisionStrong proficiency in Python PyTorchHands-on experience training object detection models (e.g., YOLO-style / Faster R-CNN / transformer-based detectors, etc.)Solid software engineering skills in a Linux environmentStrong ownership mindset: able to maintain and advance the full ML stack end-to-endMotivated to learn and apply new methods and improve production qualityMust HaveNative German speaker (customers and field partners are Germany-based)Nice to HaveExperience deploying ML models to edge devices / embedded environmentsFamiliarity with performance profiling / inference optimizationExperience with real-time video pipelines and production CV systemsWhat’s On OfferHybrid working model in MunichFlat hierarchies, high ownership, hands-on cultureInternational, multicultural environment with colleagues across multiple regionsDirect impact on a product entering real-world rollout with major German retailersBenefits/perks including mobility options, company events, and additional corporate benefits

Senior / Principal Research Scientist – Core AI Algorithms (Autonomous Systems)Location: Germany (Remote-first within Germany, on-site in Frankfurt every 2–4 weeks)About the RoleWe are partnering with a global automotive OEM building a core AI research and algorithm team responsible for the foundational intelligence behind next-generation automated driving systems.This role is research-driven and sits upstream of product teams. The focus is on inventing, validating, and transitioning new perception and world-modeling algorithms from research into production-ready systems. The team operates similarly to a big-tech research lab, but with a clear path to real-world deployment.Research Focus AreasDepending on background and interest, you may work on topics such as:3D scene understanding and world modelingOccupancy, motion forecasting, and dynamic scene reconstructionMulti-sensor perception (camera, LiDAR, radar)Representation learning for autonomous systems (BEV, implicit / generative 3D, Gaussian models, foundation models)Robustness, generalization, and long-tail perceptionLearning under weak, sparse, or noisy supervisionBridging offline training with real-world deployment constraintsKey ResponsibilitiesConduct original research in perception and autonomous systems with clear technical ownershipDesign and prototype novel algorithms and learning frameworksPublish at or contribute toward top-tier conferences and journals (e.g., CVPR, ICCV, ECCV, NeurIPS, ICRA, IROS)Translate research ideas into scalable, production-oriented implementationsCollaborate with applied ML, systems, and hardware teams to ensure feasibilityShape the long-term technical roadmap of the core AI organizationMentor junior researchers and engineers where appropriateRequired BackgroundPhD (or equivalent research experience) in Computer Vision, Machine Learning, Robotics, or a related fieldStrong publication record at top-tier conferences or journalsExperience conducting research within an industrial or applied settingExcellent understanding of modern deep learning methods and 3D perceptionStrong programming skills in Python and/or C Ability to work across the full spectrum from theory to implementationStrongly PreferredResearch experience in autonomous driving, robotics, or embodied AIWork on 3D perception, tracking, SLAM, or world modelsExperience at big-tech research labs, industrial AI labs, or advanced OEM R&DFamiliarity with real-world constraints such as runtime, memory, and system integrationPrior collaboration with product or engineering teamsWhat’s on OfferA research-first role with real influence on production systemsThe opportunity to define core algorithms, not just incremental improvementsA team culture that values publications, patents, and long-term thinkingRemote-first working model within Germany, with regular in-person collaboration in FrankfurtCompetitive compensation aligned with senior / principal research profilesWho This Role Is ForResearchers who want their work to ship into real vehiclesIndustry researchers seeking greater technical ownershipPhD-level candidates who enjoy both publishing and buildingProfiles combining academic depth with practical engineering maturityLooking forward to seeing your profile!

Location: Europe (strong preference for Spain, ideally Madrid) Type: Full-time About the Company We're working with a high-growth startup developing AI systems that allow industrial robots to perform tasks they currently cannot, starting with complex warehouse operations like mixed palletizing. Their technology combines deep reinforcement learning (DRL) with modern sequence modeling to tackle control and combinatorial optimization problems where classical approaches fail.They are a small, highly skilled team. Joining us means having direct impact, minimal bureaucracy, and ownership over core technology that will be deployed in real-world, high-throughput environments. Role Overview As the second hire in the DRL team, you will own the end-to-end reinforcement learning stack: from problem formulation to algorithm design, large-scale training, evaluation, and deployment. You will work closely with the technical leadership to translate cutting-edge DRL research into practical production throughput at operational sites. This role is highly autonomous, requiring a hands-on expert capable of leading experiments, troubleshooting complex issues, and establishing best practices for algorithm development and deployment. Key ResponsibilitiesDesign, implement, and ship DRL algorithms (e.g., PPO, SAC, DDQN and variants) incorporating advanced architectures such as encoders, cross-attention, and pointer networksOptimize stability and sample efficiency using techniques such as GAE, reward shaping, normalization, entropy/KL control, curriculum learning, and distributional/value-loss tuningSet up and manage large-scale training pipelines: multi-GPU training, parallel rollouts, efficient replay/storage, reproducible experimentsProductionize algorithms with clean, maintainable PyTorch code, profiling, Dockerized services, cloud deployments (AWS), experiment tracking, and dashboardsCollaborate with leadership to align technology with business goals and customer needsMentor and grow future team members, fostering a culture of technical excellence and innovationRequired QualificationsProven track record delivering DRL systems beyond academic demos: led at least one end-to-end DRL system from concept to production or achieved a state-of-the-art benchmark in the last 3–5 yearsDeep expertise in reinforcement learning and deep learning, with strong PyTorch skillsSolid understanding of DRL theory: MDPs, Bellman operators, policy gradients, trust-region/KL methods, λ-returns, stability and regularization in on-policy/off-policy regimesSystems experience: Python, Linux, multi-GPU training, Docker, cloud deployments (AWS preferred)Comfortable taking ownership of experiments, code quality, and results in a small, high-impact teamPhD or equivalent experience in DRL is acceptable; strong academic-only candidates considered if they demonstrate deep expertiseNice to HaveRobotics experience is not requiredProduction system deployment experience is beneficial but not mandatoryLocation & TravelEU-based (CET ±1) with occasional travel to customer sitesPreference for candidates in Spain; otherwise, EuropeCompetitive Compensation & Real Equity Offered.Interview ProcessDeep Technical Session – with CTO, focused on past DRL work (no coding tests, no homework)Traits & Skills Interviews – Two × 1-hour sessions with co-founders to assess problem-solving, communication, and startup fitTeam Meet & Offer – final discussion and reference checkWhy This Role is ExcitingWork at the frontier of DRL robotics in real-world, high-throughput industrial applicationsHigh autonomy, technical ownership, and direct impact on deployed AI systemsSmall, experienced founding team and strong early customer traction reduces commercial risk while maximizing technical challengeOpportunity to join a founding-stage team with equity and influence over core product and technology

Senior SLAM Engineer Location: Zurich Type: Full-time, On-site Company Overview Our client is an early-stage robotics company developing autonomy and intelligent assistance systems for large-scale mobile machinery. By combining learning-based automation with advanced remote operation, thier technology enables a single operator to safely supervise and control multiple machines in complex, real-world environments. The team brings deep academic and industrial expertise in large-scale robotics and perception, and is focused on transitioning state-of-the-art research into production systems deployed on real machines operating in demanding conditions. Role Overview This role sits at the intersection of perception, state estimation, and real-world deployment. You will contribute to the design, implementation, and deployment of advanced localization and mapping solutions for autonomous and semi-autonomous heavy machines. The systems you work on integrate multiple sensing modalities—spanning lidar, vision, inertial sensing, and satellite positioning—into a hardware-agnostic autonomy stack that can be adapted to a wide range of machine types and vintages. The role requires not only strong algorithmic expertise, but also a focus on production-quality software and system robustness. Key ResponsibilitiesDesign, prototype, and deploy real-time localization, mapping, state estimation, and calibration algorithms for large autonomous mobile platformsDevelop SLAM pipelines leveraging lidar, inertial, visual, and GNSS data sourcesOptimize system performance, robustness, and reliability under real-world operating conditionsDefine and maintain testing procedures, validation strategies, and performance metricsCollaborate closely with engineers across perception, controls, systems, and hardware to improve end-to-end autonomy performanceEnsure high-quality, maintainable implementations suitable for deployment on production systemsRequired QualificationsMaster’s or PhD in Computer Science, Robotics, Electrical Engineering, Mechanical Engineering, or a related field3 years of hands-on experience developing and deploying localization and mapping systemsStrong experience implementing SLAM and state estimation algorithms using lidar-inertial-visual sensor fusionProficiency in C and Python, with a focus on production-grade software developmentExperience working in Linux-based development environmentsAbility to manage technical risk, re-prioritize work, and meet deadlines in a fast-paced engineering environmentStrong communication skills, with the ability to explain complex technical concepts to both technical and non-technical audiencesNice to HaveExperience integrating RADAR and/or GPS/GNSS into localization or SLAM systemsFamiliarity with ROS2 and modern robotics middleware

Senior / Principal Research Scientist – Core AI Algorithms (Autonomous Systems)Location: Germany (Remote-first within Germany, on-site in Frankfurt every 2–4 weeks)About the RoleWe are partnering with a global automotive OEM building a core AI research and algorithm team responsible for the foundational intelligence behind next-generation automated driving systems.This role is research-driven and sits upstream of product teams. The focus is on inventing, validating, and transitioning new perception and world-modeling algorithms from research into production-ready systems. The team operates similarly to a big-tech research lab, but with a clear path to real-world deployment.Research Focus AreasDepending on background and interest, you may work on topics such as:3D scene understanding and world modelingOccupancy, motion forecasting, and dynamic scene reconstructionMulti-sensor perception (camera, LiDAR, radar)Representation learning for autonomous systems (BEV, implicit / generative 3D, Gaussian models, foundation models)Robustness, generalization, and long-tail perceptionLearning under weak, sparse, or noisy supervisionBridging offline training with real-world deployment constraintsKey ResponsibilitiesConduct original research in perception and autonomous systems with clear technical ownershipDesign and prototype novel algorithms and learning frameworksPublish at or contribute toward top-tier conferences and journals (e.g., CVPR, ICCV, ECCV, NeurIPS, ICRA, IROS)Translate research ideas into scalable, production-oriented implementationsCollaborate with applied ML, systems, and hardware teams to ensure feasibilityShape the long-term technical roadmap of the core AI organizationMentor junior researchers and engineers where appropriateRequired BackgroundPhD (or equivalent research experience) in Computer Vision, Machine Learning, Robotics, or a related fieldStrong publication record at top-tier conferences or journalsExperience conducting research within an industrial or applied settingExcellent understanding of modern deep learning methods and 3D perceptionStrong programming skills in Python and/or C Ability to work across the full spectrum from theory to implementationStrongly PreferredResearch experience in autonomous driving, robotics, or embodied AIWork on 3D perception, tracking, SLAM, or world modelsExperience at big-tech research labs, industrial AI labs, or advanced OEM R&DFamiliarity with real-world constraints such as runtime, memory, and system integrationPrior collaboration with product or engineering teamsWhat’s on OfferA research-first role with real influence on production systemsThe opportunity to define core algorithms, not just incremental improvementsA team culture that values publications, patents, and long-term thinkingRemote-first working model within Germany, with regular in-person collaboration in FrankfurtCompetitive compensation aligned with senior / principal research profilesWho This Role Is ForResearchers who want their work to ship into real vehiclesIndustry researchers seeking greater technical ownershipPhD-level candidates who enjoy both publishing and buildingProfiles combining academic depth with practical engineering maturityLooking forward to seeing your profile!

Deep Learning Engineer – Advanced Robotics & VLM/VLALocation: Flexible / Remote (UK or Europe preferred) Employment Type: Full-timeAbout the CompanyOur client is an ambitious AI and robotics company developing next-generation humanoid systems designed to transform how intelligent automation supports industrial and everyday environments. Their mission is to advance human potential through robotics that are scalable, safe, and capable of performing complex real-world tasks. This is a rare opportunity to work at the intersection of deep learning, multimodal AI, and robotic embodiment, helping shape the foundations of a truly intelligent automation platform. The Role As a Deep Learning Engineer, you’ll design and train large-scale models that power robotic control and perception — from foundational representation learning to behaviour cloning and reinforcement learning. You’ll work across the full data-to-deployment lifecycle, experimenting with cutting-edge multimodal architectures and building robust pipelines for high-performance, real-time systems. Key ResponsibilitiesDevelop and train deep learning models for manipulation, navigation, and general policy learning.Collaborate with teleoperations and simulation teams to define data collection goals and bridge the sim-to-real gap.Train and fine-tune multimodal LLMs, VLMs, and VLAs, integrating diverse sensory modalities (vision, audio, proprioception, LiDAR, etc.).Build scalable data pipelines for continuous ingestion, curation, weak supervision, and retraining.Partner with MLOps and infrastructure teams to enable distributed training and optimize models for real-time deployment.Contribute to shaping the next generation of embodied AI systems for safe, efficient automation.About You3 years of experience building and deploying deep learning systems (industry or research).Strong proficiency in Python and PyTorch or JAX.Hands-on experience with LLMs, VLMs, or generative models for image/video.Deep understanding of training infrastructure (streaming datasets, checkpointing, distributed compute).Strong communicator with clear experiment documentation and the ability to explain complex technical decisions.Bonus PointsExperience in robotics, autonomous driving, or other embodied AI domains.Background in reinforcement learning (PPO, DPO, SAC, etc.) or RL for LLMs.Experience optimizing deep nets for production (latency, telemetry, on-device inference).Publications at top-tier ML conferences (ICLR, NeurIPS, ICML) or significant open-source contributions.Familiarity with OpenVLA, π models, or similar embodied AI frameworks.What’s on OfferCompetitive compensation including stock options.Flexible remote-first setup with opportunities for international collaboration.Work with world-class researchers and engineers building truly transformative technology.A fast-paced, innovation-driven culture where ideas move quickly from concept to prototype.

We are looking for an experienced Engineer/Scientist to join a cutting-edge robotics team in Zurich, focused on humanoid manipulation through state-of-the-art AI. You will contribute to developing and deploying learning-based controllers to enable robots to interact intelligently, reliably, and flexibly with complex environments. This role is deeply hands-on with real robot hardware, leveraging your expertise across vision-language-action (VLA) models, diffusion models, reinforcement learning, and dexterous manipulation. What we’re looking for:PhD or Master's in Robotics, AI, or related fieldsStrong expertise in learning-based dexterous manipulation (multi-fingered grasping, task-oriented manipulation)Practical experience with running real-time controllers on robotic hardwareDeep knowledge of diffusion models, transformers, CVAEs, normalizing flowsExperience working with ROS, PyTorch, and PythonFamiliarity with vision-language models (VLA/VLMs/LLMs) for robotic planning or failure detectionBonus: experience with synthetic data generation, simulation environments, teleoperation systemsWhy this role?Join a high-calibre team (ex-ETH / ex-NVIDIA founders) backed by significant investment (Seed Series A secured)Work at the intersection of simulation, reinforcement learning, and robotics hardwareContribute to next-generation humanoid robots without reliance on traditional imitation learning paradigmsSignificant stock offering.If you’re passionate about building more intelligent robots through cutting-edge AI and want to work in a fast-moving, well-resourced environment, we’d love to hear from you.