Sort by
Refine Your Search
-
efforts across scientific systems. Develop and apply federated learning on distributed and heterogenous datasets. Develop more efficient and resilient DP techniques that minimize performance loss while
-
circuit design, power distribution hardware, embedded systems, and industrial sensing applications. The successful candidate will be a highly motivated and creative problem solver, comfortable parsing
-
for sponsors across the Federal Government. We're looking for staff who bring fresh ideas from areas including information retrieval, distributed computing and storage, and data processing workflows. We need our
-
, independent judgment. task leading, and document development. Work may include power distribution systems, electrical design services for beamline instruments and conventional facilities, and experimental
-
of experience in facility systems, operations, and maintenance. Preferred Qualifications: Experience in HVAC and their control systems, industrial piping design, and in 480 VAC 3 phase power distribution
-
for Science @ Scale: Pretraining, instruction tuning, continued pretraining, Mixture-of-Experts; distributed training/inference (FSDP, DeepSpeed, Megatron-LM, tensor/sequence parallelism); scalable evaluation
-
electrical systems, UPS, HVAC, cooling systems, and power distribution in data center environments. Familiarity with Data Center Infrastructure Management (DCIM) tools including telemetry ingestion and
-
alignment: Implement efficient partitioning, indexing, and distributed processing strategies while coordinating with engineers and scientists to align infrastructure with geospatial workflows and analytical
-
across the Federal Government. We are looking for engineers who bring strong software engineering fundamentals and fresh ideas from areas such as distributed computing, information retrieval, scalable
-
for sponsors across the Federal Government. We're looking for staff who bring fresh ideas from areas including information retrieval, distributed computing and storage, and data processing workflows. We need our