Member Of Technical Staff - Ml Performance
Veeda AI
About Us
Veeda AI is building the next generation of multimodal foundation world models for Physical AI. We're a small, fast-moving team of engineers and researchers from leading AI labs, tackling some of the most challenging problems at the intersection of AI, robotics, and embodied intelligence. If you're excited about pushing the boundaries of what's possible with Physical AI, you'll have the opportunity to make an outsized impact from day one.Responsibilities
Distributed Training Throughput: Own step time and model FLOPs utilization for multi-node video world model training, choosing the tensor, context, and expert parallelism mix in PyTorch FSDP2 and Megatron-Core rather than inheriting a default.Inference Throughput Optimization: Profile end-to-end latency, GPU utilization, memory pressure, and kernel efficiency with PyTorch Profiler, Nsight Systems, Nsight Compute, and torch.Utils.Benchmark, then improve throughput through torch.Compile/Inductor, CUDA Graphs, mixed and low precision, quantization, operator fusion, and multi-GPU serving.Precision & Numerical Stability: Take BF16, FP8, and NVFP4 recipes from running to converging on Blackwell, chasing scaling-factor and accumulation bugs into the video tokenizer and VAE layers where the activation outliers actually live.Kernels & Compilation: Write and tune the CUDA and Triton kernels PyTorch does not give us, driving FlashAttention-4, FlexAttention, and torch.Compile integration so quadratic attention over long video sequences stops setting step time.Communication & Overlap: Tune NCCL collectives and compute/communication overlap across NVLink domains and the fabric, using the NCCL flight recorder to turn a watchdog timeout into a named rank and collective, not a restart.Fault Diagnostics & Recovery: Build the detection layer for silent data corruption (SDC), stuck CUDA kernels, and\ "card-freeze\" hangs, plus asynchronous and tiered checkpointing that makes an interruption cost minutes rather than a day.Requirements
Bachelor's degree or equivalent hands-on experience in Computer Science, Computer Engineering, or a related technical field.Deep hands-on experience with PyTorch and at least one large-scale parallelism stack (FSDP2, Megatron-Core, TorchTitan, or DeepSpeed) on real multi-node jobs, not single-node approximations.Fluency in Python and C++/CUDA with the ability to predict where a kernel will stall from its memory access pattern before profiling.Experience profiling live training and inference runs with Nsight Systems or the PyTorch profiler and translating traces into quantifiable step-time or MFU improvements.Expertise in at least one of low-precision numerics, kernel authoring, or large-run fault diagnosis, and credibility in the others.Nice to Have
Experience optimizing training or inference workloads across very large GPU clusters, including topology-aware placement, scaling efficiency, performance isolation, and diagnosing failures that emerge only at fleet scale.Experience writing Triton, CUTLASS, or CuTe-DSL kernels, or contributing to open-source kernel libraries.Experience implementing context or sequence parallelism for long-horizon video or high-token-count models.Experience running or porting large training workloads on AMD GPUs (ROCm) or Google TPUs (JAX/XLA).Experience building fault-tolerant training with elastic world size, dynamic node re-queueing, or asynchronous distributed checkpointing.Experience optimizing generative inference for interactive rollouts, including few-step samplers, distillation, and KV or latent caching.Publications or presentations on machine learning systems, compilers, or high-performance kernels.
#J-18808-Ljbffr
Veeda AI is building the next generation of multimodal foundation world models for Physical AI. We're a small, fast-moving team of engineers and researchers from leading AI labs, tackling some of the most challenging problems at the intersection of AI, robotics, and embodied intelligence. If you're excited about pushing the boundaries of what's possible with Physical AI, you'll have the opportunity to make an outsized impact from day one.Responsibilities
Distributed Training Throughput: Own step time and model FLOPs utilization for multi-node video world model training, choosing the tensor, context, and expert parallelism mix in PyTorch FSDP2 and Megatron-Core rather than inheriting a default.Inference Throughput Optimization: Profile end-to-end latency, GPU utilization, memory pressure, and kernel efficiency with PyTorch Profiler, Nsight Systems, Nsight Compute, and torch.Utils.Benchmark, then improve throughput through torch.Compile/Inductor, CUDA Graphs, mixed and low precision, quantization, operator fusion, and multi-GPU serving.Precision & Numerical Stability: Take BF16, FP8, and NVFP4 recipes from running to converging on Blackwell, chasing scaling-factor and accumulation bugs into the video tokenizer and VAE layers where the activation outliers actually live.Kernels & Compilation: Write and tune the CUDA and Triton kernels PyTorch does not give us, driving FlashAttention-4, FlexAttention, and torch.Compile integration so quadratic attention over long video sequences stops setting step time.Communication & Overlap: Tune NCCL collectives and compute/communication overlap across NVLink domains and the fabric, using the NCCL flight recorder to turn a watchdog timeout into a named rank and collective, not a restart.Fault Diagnostics & Recovery: Build the detection layer for silent data corruption (SDC), stuck CUDA kernels, and\ "card-freeze\" hangs, plus asynchronous and tiered checkpointing that makes an interruption cost minutes rather than a day.Requirements
Bachelor's degree or equivalent hands-on experience in Computer Science, Computer Engineering, or a related technical field.Deep hands-on experience with PyTorch and at least one large-scale parallelism stack (FSDP2, Megatron-Core, TorchTitan, or DeepSpeed) on real multi-node jobs, not single-node approximations.Fluency in Python and C++/CUDA with the ability to predict where a kernel will stall from its memory access pattern before profiling.Experience profiling live training and inference runs with Nsight Systems or the PyTorch profiler and translating traces into quantifiable step-time or MFU improvements.Expertise in at least one of low-precision numerics, kernel authoring, or large-run fault diagnosis, and credibility in the others.Nice to Have
Experience optimizing training or inference workloads across very large GPU clusters, including topology-aware placement, scaling efficiency, performance isolation, and diagnosing failures that emerge only at fleet scale.Experience writing Triton, CUTLASS, or CuTe-DSL kernels, or contributing to open-source kernel libraries.Experience implementing context or sequence parallelism for long-horizon video or high-token-count models.Experience running or porting large training workloads on AMD GPUs (ROCm) or Google TPUs (JAX/XLA).Experience building fault-tolerant training with elastic world size, dynamic node re-queueing, or asynchronous distributed checkpointing.Experience optimizing generative inference for interactive rollouts, including few-step samplers, distillation, and KV or latent caching.Publications or presentations on machine learning systems, compilers, or high-performance kernels.
#J-18808-Ljbffr
Vacancy posted 1 hour ago
Similar jobs that could be interesting for youBased on the Member Of Technical Staff - Ml Performance in Winnipeg, MB vacancy
- ...impact from day one. Responsibilities ML Methods for Data: Design, develop, and... ...data quality and assess its impact on model performance. Iterative Improvement: Use failure... ...Computer Science, Engineering, or a related technical field, or equivalent hands-on experience....Performance
$250k per year
...Montreal, Paris, Berlin and Seoul. Joinus!Role Overview As a Performance Engineer in the Pre-Training team you will be responsible foroptimizing... ...software engineering skills.Proficiency in Python and related ML frameworks such as JAX, Pytorch andXLA/MLIR.Experience writing...PerformanceFull timeWork at officeLocal areaRemote workHome office- ...laboratories. We create high-performance airflow control solutions that... ...precise airflow control. Staff Accountant – Antec Controls... ...customers succeed. We combine technical expertise, accountability, and... ..., persons with disabilities, members of visible minorities, individuals...PerformanceWorldwide
$220k - $405k per year
...implementing, and scaling these interfaces. As a member of our team, you'll work on an eclectic portfolio spanning distributed systems, performance optimization, agent orchestration, and... ...experience alike. You'll also define technical strategy for how we scale to meet...PerformanceContract work- ...Technical Lead – Java & Enterprise Integration Information Technology Who we are ML arteka is one of the fastest-growing IT Services and... ...We are looking for a team members who embody our values: Empathetic... ...balancing, mentoring, performance feedback, and day-to-day...PerformanceContract workFlexible hours
- ...CTO. We're based in St. John's, Newfoundland, with remote team members in Nova Scotia, New Brunswick, and Ontario. Four times a year, the... ...SiftMed might be the fit for you You balance pragmatism with technical excellence, and you know which one a problem calls for You're...InternshipWork at officeRemote workFlexible hours
- ...tangible, lasting benefits to students, faculty, staff, and local communities. Navitas is seeking... ...Duties Relieve/assist the other members of the finance team as required Ensure... ...internal and external customer service Perform other duties as required Essential Skills...Full timeInternshipWork at officeLocal area
$160k - $210k per year
...Continuum. We're looking for a staff software engineer to help... ...Take part in a collaborative, technically rigorous culture, and help shape... ...3 months, you’ll be a full member of the team, contributing to... ...software products and lean, high‑performing teams by unlocking the...InternshipWork at officeLocal areaRemote workWorldwideFlexible hours$140k - $240k per year
...) On-site | Full-time Compensation: $140K - $240KThe Role: Member of Technical Staff (Applied AI) The Senior Applied AI Engineer will play a pivotal... ...the integration of AI agents with browser environments.Performance Benchmarking: Design and implement evaluation frameworks to...PerformanceFull timeInternshipWork at office- ...systems work at scale. This role demands deep technical expertise in production environments and... .... Comfortable diving into complex ML codebases to identify and resolve issues,... ...responsibilities: Design and write high-performing and scalable software for training models...Full timeWork at officeLocal areaRemote workHome office
$250k per year
...group of researchers and engineers focused on building reliable ML systems and pushing the boundaries of LLM inference efficiency. We... ...this team, you’ll work across the inference stack to improve core performance metrics by diving deep into model execution, identifying...PerformanceFull timeWork at officeLocal areaRemote workHome office- ...results, and iterating on models to optimize performance. Participate in swift iteration cycles... ...to understand requirements and provide technical guidance. What We're Looking For (... ...meet product needs. Experience deploying ML models in the Cloud. Prior experience...PerformanceRemote workNight shift
- ...AI systems work at scale. This role demands deep technical expertise in production environments and the... ...talent to ensure you can do your best work. As a Member of Technical Staff, you will: Design and write high‑performing and scalable software for training models....Full timeWork at officeLocal areaRemote workHome office
- ...City: Winnipeg What you will do: Cast Member Cineplex’s Cast Members are an integral part of the team who provide an exceptional entertainment experience to each and every guest. Cast Members have a range of responsibilities, each of which help the theatre’s daily...InternshipSeasonal workWork at officeFlexible hoursShift workNight shiftAfternoon shift
- ...Requirements Bachelor's degree or equivalent hands‑on experience in Computer Science, Computer Graphics, Robotics, or a related technical field. Experience building or substantially extending a physics engine or simulation framework (Newton, MuJoCo, PhysX, or something...
- Enhance AI capabilities at Onix as a Member of Technical Staff with a focus on small language models. Engage in innovative projects combining efficient... ...rapid prototyping Requirements: • Significant experience in ML training and evaluation • Proven skills in distillation and...
- Contribute to groundbreaking Physical AI innovations at Veeda AI as a Technical Staff Member. Engage in multi-node video world model training while tackling key performance challenges. Veeda AI is seeking a Technical Staff Member to enhance its world-class team working with...Performance
- ...new ones. We prioritize the thorough training of all team members in every aspect of our back-of-house operations. This collaborative... ...Hours Advancement Opportunities Benefit packages Performance based bonuses Staff Discount Lifelong Friendships When not at work, you...PerformanceLocal areaFlexible hours
- ...Chez Ubisoft, programmez des systèmes ML en C++ avec une expertise en moteurs d’inférence, tests, performance et architecture logicielle. Ce poste de programmeur principal C++ contribue à la feuille de route des applications d’apprentissage automatique. Vous développerez...PerformanceApprenticeship
- Senior Staff ML Compiler and Performance EngineerMarkham, Ontario Senior Staff Compiler EngineerCanada - RemoteWe are partnered with a global leader... ...this a reality. Job Purpose & Responsibilities As a member of the ML Systems Team, you will participate in two...Performance
- ...incumbent provides therapeutic, diagnostic, technical and teaching respiratory therapy services... ...care. Ability to stand and walk and perform manual ventilation with one or both hands... ...aggressive and/or agitated clients/staff/visitors. St. Boniface Hospital is committed...Shift workNight shift
- ...to join the Model Efficiency team, focusing on building reliable ML systems and optimizing LLM inference. You will improve the inference... ...across global offices. Ideal candidates have 5+ years of high-performance coding experience in C++ or Python and familiarity with LLM ecosystems...PerformanceRemote work
- ...Fortune 100 companies. Our work covers audio, video, and other new data types that advance perceptual intelligence. We are hiring Members of Technical Staff to create new data infrastructure as well as improve upon existing infrastructure. The Role We’re looking for an applied...Long term contractTemporary workWork at office
$155k - $230k per year
...manufacturers and are working with over 50+ pharma brands. Frontiers of technology: We're building applications and rapidly iterating technical approaches at the frontiers of AI, experimenting with LLMs and new frameworks. We are incorporating new, advanced frameworks to our...Work at officeRemote workNight shift- ...Pharmacists license a must Superior customer relations skills and a strong sense of professionalism Excellent leadership and performance management skills required Proven ability to build the business using community based marketing strategies Excellent communication...PerformanceHourly payPermanent employmentTemporary workInternshipImmediate start
- Member of Technical Staff, Small Language ModelsFull-time Onix is the Personal Intelligence platform. Each onix is a small language model trained... ...experts directly when their voice, fidelity, or persona needs ML‑level attention. What You Will Work With Training stack...Live InRelocation
- ...Senior Evaluation and Performance Analyst Department: Executive and Indigenous Affairs This Job Opening is limited to employees... ...projects while collaborating with partners and mentoring junior team members. Key Qualifications: Designation as a Credentialed...PerformanceLong term contract
- ...customers use Java, bringing them the tools and insight they need to deliver faster, more performant, more dependable, and less expensive Java services. Come join one of the most influential members of the Java and OpenJDK community and help influence the future of Java. Falcon...PerformanceRemote workWorldwideFlexible hours
$27 per hour
...the World's Most Trustworthy Companies in 2026. About Channel Performance: The Channel Performance Service Line is home to the largest... ...-matter experts from around the world, combining thematic and technical experts to deliver top-quality research and insights. Simply speaking...PerformanceHourly payPermanent employmentFor contractorsSecond jobRemote workWorldwideMonday to fridayFlexible hours- ...incumbent provides therapeutic, diagnostic, technical and teaching respiratory therapy services... ...care. Ability to stand and walk and perform manual ventilation with one or both hands... ...aggressive and/or agitated clients/staff/visitors. St. Boniface Hospital is committed...Immediate startShift workNight shift
Do you want to receive more vacancies?
Subscribe and receive similar vacancies to Member Of Technical Staff - Ml Performance. Be the first to apply!
Related searches
- service support technician Winnipeg, MB
- sr. IT support analyst Winnipeg, MB
- technical analyst Winnipeg, MB
- technical system analyst Winnipeg, MB
- desktop support analyst Winnipeg, MB
- senior technical analyst Winnipeg, MB
- network technical analyst Winnipeg, MB
- service desk analyst Winnipeg, MB
- senior technical specialist Winnipeg, MB
- decision support analyst Winnipeg, MB

