5+ years of experience in building and deploying ML systems, particularly in computer vision.
Deep expertise in object detection and segmentation, especially with YOLO and DETR models.
Hands-on experience with vision-language models (VLMs) and their fine-tuning.
Familiarity with vision-language agents (VLAs) or related multimodal learning.
Strong proficiency in PyTorch and experience with distributed training and GPU profiling.
Proven ability to transition models from research to production, including optimization and monitoring.
Demonstrated leadership in cross-team projects and effective communication with diverse stakeholders.
Responsibilities
Define technical strategy and architecture for vision models across various workloads.
Train and adapt models using proprietary data, focusing on techniques like distillation and quantization.
Develop data engines, including labeling pipelines and active learning systems.
Design and implement rigorous evaluation metrics and benchmarks for model performance.
Deploy models to production while adhering to latency and cost constraints.
Lead technical projects, mentor engineers, and contribute to design documentation.
Stay updated on research trends and integrate relevant findings into prototypes.
Benefits
Drive operational excellence in a rapidly growing AI company.
Shape the future of AI project delivery with scalable processes.
Collaborate with top engineering talent to enhance team performance.
Gain exposure to innovative Vision AI and IoT solutions in manufacturing.
Lead initiatives that significantly impact company growth and customer satisfaction.
Build expertise in the evolving field of industrial AI and autonomous manufacturing.
Full Job Description
What you'll do
Define the technical strategy and architecture for vision models across detection, VLM, and VLA workloads.
Train, fine-tune, and adapt models (YOLO variants, open VLMs, VLA policies) on proprietary data, including distillation, quantization, and LoRA-style adaptation.
Build data engines: labeling pipelines, auto-labeling with foundation models, active learning, and synthetic or simulated data.
Design rigorous evaluation: benchmarks, regression suites, failure-mode analysis, and online metrics tied to product outcomes.
Ship models to production under latency, memory, and cost limits, on cloud GPUs and edge hardware (TensorRT, ONNX, Jetson or similar).
Lead cross-team technical projects; write design docs; review code and models; mentor senior and mid-level engineers.
Track the research frontier, decide what is worth adopting, and turn it into prototypes and product.
Requirements What we're looking for
5+ years building and shipping ML systems, with focus on computer vision; BS/MS/PhD in CS, EE, or a related field, or equivalent experience.
Deep hands-on experience with object detection and segmentation (YOLO family, DETR-style models), including training, tuning, and production deployment.
Hands-on experience with VLMs (for example LLaVA-, Qwen-VL-, or PaliGemma-style models): fine-tuning, prompting, grounding, and evaluation.
Experience with VLAs or embodied and robot-learning models (for example OpenVLA-, RT-2-, or pi0-style policies), or a clear record in adjacent multimodal work.
Strong PyTorch skills and fluency with distributed training, mixed precision, and GPU profiling.
Track record of taking models from research to production: model optimization, inference serving, monitoring, and iteration on real-world failures.
Evidence of staff-level impact: setting direction across teams, making sound build-versus-adopt calls, and raising the bar for others.
Clear written and verbal communication with engineers, researchers, and product leaders.
Nice to have
Publications or open-source contributions in vision or multimodal learning.
Experience with video understanding, 3D perception, or multi-sensor fusion.
Experience with simulation, sim-to-real transfer, or reinforcement and imitation learning.
Edge and embedded deployment experience (TensorRT, CUDA, quantization-aware training).
Experience with LLM tooling such as vLLM, or with RLHF-style post-training for multimodal models.
This role may support work subject to U.S. export controls (ITAR). Access to certain projects may require U.S.-person status (U.S. citizen or national, lawful permanent resident, refugee, or asylee) or export authorization.
Benefits Why Join Maneva?
Drive operational excellence in a fast-growing AI company transforming manufacturing through cutting-edge vision systems
Shape the future of project delivery by building scalable processes for AI system deployments
Work with world-class engineering talent while optimizing team performance and customer success
Gain exposure to cutting-edge Vision AI & IoT solutions across diverse manufacturing environments
Lead strategic initiatives that directly impact company growth and customer satisfaction
Build expertise in the rapidly expanding field of industrial AI and autonomous manufacturing
Compensation & Benefits
Salary based on location and experience - market rate
Equity participation in a high-growth AI startup
Comprehensive health, dental, and vision coverage
Flexible PTO and professional development opportunities
Remote-first culture with modern collaboration tools
