ExpertiseUpdated on 6 February 2026
HPC Services for Distributed Systems
Head of innovation at UST
Barcelona / Madrid, Spain
About
HPC Services for Distributed Systems
UST provides high-performance computing (HPC) expertise for large-scale, distributed workloads across sectors. We bridge academic research (Stanford SAIL, MIT CSAIL) with enterprise-grade infrastructure and industrial applications.
What we bring:
-
Dedicated H100 on-prem cluster for training, inference, and experimentation
-
Production MLOps/LLMOps pipelines on Azure, AWS, GCP
-
Familiarity with MGHPCC and Mass Open Cloud Alliance environments for HPC-tier workloads
-
Cross-tier orchestration: policy-aware query routing, carbon- and cost-aware scheduling, long-context memory across edge→cloud→HPC
Work areas:
-
Continuum routing & scheduling: PAQR reference implementation, adapters for vLLM/Triton/Bedrock, dashboards
-
Distributed RAG & long-context orchestration: KV offload protocols, semantic caching, student–expert distillation
-
Industrial pilots leveraging HPC tier: healthcare, retail, and financial services
TRL path: Year 1: prototypes & benchmarks; Year 2: integrated pilot stacks; Year 3: hardened demonstrators, third-party evaluation.
See also our expertise in AI services, including agentic automation, document AI, and privacy-preserving analytics.
Similar opportunities
Expertise
AI Services for Distributed Systems
Daniel Field
Head of innovation at UST
Barcelona / Madrid, Spain
Expertise
Acube – Governing AI from the Mathematical Core
- Arificial Intelligence and Data
- Business models and exploitation strategies
Fabrizio Coccetti
Project Manager at Acube srl
Teramo, Italy
Expertise
Vision AI and Industrial AI Infrastructure for Inline Inspection and Data-Driven Manufacturing
- Arificial Intelligence and Data
- Testbeds, pilot lines, demonstrations
- Modelling, simulations and digitally aided design
Özgecan Sarı
Product Manager @robolaunch at robolaunch
Ankara, Türkiye