Top Tech Jobs & Startup Jobs

Design and adapt vision-language and video models for scene understanding, temporal reasoning, and action recognition. Build GPU-based training and evaluation pipelines, curate video-text datasets, develop benchmarks, optimize models through compression and distillation, and deploy production-ready inference systems. Collaborate with computer vision, platform, robotics, product, and customer teams to translate research into usable AI products.

Deploys and supports AI edge systems at industrial customer sites. Responsibilities include installing cameras, compute nodes, networking equipment, and hardware cabinets; configuring devices; validating connectivity and edge-to-cloud data flow; automating installation and diagnostics with scripts; troubleshooting hardware, network, and software issues; coordinating logistics; and ensuring reliable, safe, high-performance installations.

Own customer AI onboarding from initial execution through go-live and continuous improvement. Validate camera and input data quality, identify edge cases, maintain datasets and SOPs, track success metrics, communicate blockers, and coordinate installation teams to ensure reliable, consistent deployments. Improve onboarding workflows through documentation, checklists, templates, and tool enhancement recommendations. The role requires structured execution, strong communication, process discipline, technical systems comfort, and preferably manufacturing or process experience.

Support AI software onboarding by executing onboarding cycles, checking camera and input-data quality, documenting edge cases, tracking success metrics, communicating blockers, improving SOPs and tools, and coordinating with installation teams to ensure reliable, consistent deployments.

Analyze manufacturing and production data to identify performance drivers, bottlenecks, anomalies, and optimization opportunities. Build dashboards, reports, data pipelines, features, and quality metrics while collaborating with customers, AI and backend engineers, Product, and Customer Success. Translate operational insights into improvements for customer outcomes and the company’s AI analytics platform. Occasional travel to manufacturing plants is required.

Own provisioning and operations for a distributed fleet of Linux-based edge devices, cameras, and sensors. Build PXE-based deployment and configuration automation, manage monitoring, patching, diagnostics, networking, VPNs, and firewalls, and define fleet reliability standards. Support installation engineers through deployment-ready equipment, documentation, and troubleshooting. The role is based in Berlin, involves up to 50% travel to customer sites, and may require lifting equipment weighing over 30 kg.

Lead R&D for industrial robotic manipulators, including modeling, planning, control, perception, machine learning, simulation, and hardware integration. Develop and test robotic prototypes, planning algorithms, middleware, software architectures, pose estimation, segmentation, and autonomous grasping systems. Deploy and validate solutions using ROS-compatible hardware, sensors, robot arms, and real-world environments.

Develop and maintain backend features, APIs, data pipelines, and infrastructure for a scalable platform. Improve database performance, reliability, and real-time processing; debug and optimize systems for low latency. Collaborate with AI, frontend, and product teams, contribute to architecture decisions, and take ownership of backend problems. The role also involves cloud environments, Linux performance tuning, orchestration, and hybrid edge/cloud architectures.

Support AI software onboarding for manufacturing customers by executing onboarding cycles, validating camera and input data quality, documenting edge cases, tracking success metrics, coordinating with installation teams, and improving processes, documentation, checklists, and internal tools. The role requires structured execution, clear communication, attention to detail, ownership, and comfort working with technical systems. Manufacturing or process experience is advantageous, while prior machine learning experience is not required.

Develop and deploy computer vision models for video and sensor applications, building scalable training, evaluation, inference, and MLOps pipelines. Manage GPU clusters, Docker/Kubernetes environments, cloud workloads, CI/CD, and model orchestration. Collaborate with research, robotics, and platform teams to productionize reliable computer vision systems for cloud and edge deployment, including real-time inference optimization and industrial automation applications.