21 of 21 Permanent vLLM Jobs in London

TensorFlow at a working, hands-on level. Distributed training: Horovod, DeepSpeed, Megatron-LM, or equivalent multi-node training frameworks. Inference & serving: NVIDIA Triton, vLLM, TensorRT-LLM, or equivalent high-throughput serving platforms. MLOps/LLMOps: Kubeflow, MLflow, and at least one hyperscaler ML platform (SageMaker, Azure ML, or Vertex ...

/ML services (e.g., SageMaker, Vertex AI, AzureML), including sovereign, on‐premise, and hybrid deployment models. Understanding of LLM inference stacks (vLLM, llama.cpp, OpenVINO) and model delivery formats (ONNX, safetensors, HuggingFace model hub). Understanding of benchmarking and LLM performance evaluation (accuracy, latency, throughput). Hands‐on coding skills ...

plus Model fine-tuning practical experience with LoRA/QLoRA, supervised fine-tuning, or RLHF workflows is a plus Inference optimization experience with vLLM, TensorRT-LLM, Triton, or model quantization techniques is desirable Observability tooling practical experience with LLM monitoring, tracing, and evaluation frameworks is a strong plus Familiarity with ...

React, Angular), APIs (REST/GraphQL) Machine Learning: Foundations of MLflow, Weights & Biases, LoRA, PEFT, SageMaker, Vertex AI, AzureML, Kubeflow, Airflow, Prefect, TorchServe, Triton, vLLM, SHAP, LIME, pandas, Spark, SQL LLM tooling: Familiarity with LangChain, Semantic Kernel, prompt chaining, vector search, and context management Performance & observability: Skilled in debugging distributed ...

experience with LLM training cycles, parameter-efficient fine-tuning (PEFT), and sophisticated prompt engineering. Inference Stack: Experience with high-performance inference servers (e.g., vLLM, TGI, or Triton ) and an understanding of how to optimize models for GPU deployment. Infrastructure: Comfortable working in Linux-based environments and proficient in managing containerized ...

Architectures: Hands‐on experience building with OpenAI and Anthropic models/APIs; strong knowledge of LLMs and agent architectures. Tooling: Practical experience with LiteLLM, vLLM, Langfuse (or equivalent LLM observability tooling). Experience with Claude Code/Codex and associated coding‐agent workflows is a plus. Evals: Proven ability ...

engineering & synthetic data pipelines Agentic frameworks, reasoning, tool use, memory systems Evaluation & benchmarking (LLM-as-judge, safety/alignment metrics) Inference optimisation (quantization, distillation, vLLM/TensorRT-LLM) Requirements: - PhD or Master's in CS/AI/ML/Physics/Maths or equivalent research experience - Strong hands ...

deployment pipelines and scheduling across clusters and providers. What You Bring Strong Python and system-architecture skills. Experience deploying models with PyTorch, Hugging Face, vLLM, SGLang, TensorRT-LLM, or similar. Experience with queues, scheduling, traffic control, and fleet management at scale. Experience with Linux, Docker, and Kubernetes, and with orchestration ...

APIs, SDKs, CLIs, notebooks Infrastructure: storage, networking, IAM, observability, infrastructure-as-code Familiarity with tools such as PyTorch, Ray, Slurm, MLflow, Langchain, Hugging Face, vLLM, Triton, or similar. Able to balance performance, cost, usability, flexibility, and reliability. Strong prioritisation, communication, and stakeholder management skills. Comfortable taking ambiguous problems from discovery ...

pragmatism to prioritise what matters in a fast-moving environment. Bonus: experience with GPU clusters or ML training/inference infrastructure (e.g. vLLM, Slurm), on-prem/bare-metal or deep networking experience, or open-source contributions in the cloud-native community. Location This role is remote ...

optimizing models in PyTorch or JAX Deep knowledge of at least one of these areas: LLMs: pretraining, alignment, evals GPU inference and performance: vLLM, Triton, CUDA, profiling Reinforcement learning (RL) Scientific ML: weather and climate, physics, biology Able to explain why a problem is hard for a frontier model Nice ...

transformer inference, including prefill versus decode, KV‐cache behaviour, attention variants and performance bottlenecks Hands‐on experience with LLM serving frameworks such as vLLM, SGLang, TensorRT‐LLM or TGI, and the PyTorch ecosystem Experience with inference optimisation techniques, including quantisation, speculative decoding and model parallelism across modern GPU architectures Strong ...

post-training Experience with distributed training and GPU systems Strong experimental judgement Ability to work comfortably across ML research and systems engineering Optional vLLM, SGLang, Ray, FSDP or Megatron-LM CUDA, Triton, CuTE or GPU performance optimisation Shortlisted candidates will be contacted within 48 hours. #J-18808-Ljbffr ...

throughput and cost per token, partnering with the CUDA/GPU engineers Make the core software architecture calls on serving frameworks and orchestration (e.g. vLLM, TensorRT-LLM, SGLang, Triton Inference Server or equivalents) Translate throughput, latency and uptime commitments into concrete technical specifications and serving capacity plans Act as direct ...

throughput and cost per token, partnering with the CUDA/GPU engineers. Make the core software architecture calls on serving frameworks and orchestration (e.g. vLLM, TensorRT-LLM, SGLang, Triton Inference Server, or equivalents). Translate throughput, latency, and uptime commitments into concrete technical specifications and serving capacity plans. ...

commonly used databases such as PostgreSQL, Clickhouse, and MongoDB PREFERRED SKILLS AND EXPERIENCE: Experience with LLM inference engines and serving frameworks (e.g., SGLang, TensorRT, vLLM) Experience designing or building with agent SDKs and agent orchestration frameworks Experience with Docker, Kubernetes, and containerized applications Expert knowledge of gRPC (unary, response streaming ...

generation Depth in multi‐agent systems, planning, program synthesis, or retrieval over structured artefacts such as codebases Deep familiarity with the internals of SGLang, vLLM, or comparable inference serving frameworks - scheduler design, memory management, and execution pipelines A background in another field that studies systems of interacting heterogeneous components - neuroscience ...

cost baselines against realistic customer traffic profiles, and tune deployments to hit those targets Deploy and validate new model families on inference frameworks (vLLM, SGLang), determining optimal shapes, quantization configs, and serving patterns across workloads. Model Strategy and Fine‐Tuning Guide customers on model selection, fine‐tuning strategy … issue with an ML engineer in the same afternoon. Preferred Qualifications 10+ years in technical field or engineering roles. Experience with inference serving frameworks (vLLM, SGLang, TensorRT-LLM) and tuning deployments for real workloads. Experience operating as a technical authority inside a customer's environment building within their infrastructure, navigating ...

quota management, and utilization systems across heterogeneous accelerator environments. Optimize inference pipelines for latency, throughput, reliability, memory efficiency, and cost using frameworks such as vLLM, Triton Inference Server, TensorRT, or equivalent serving stacks. Partner with ML engineers and researchers to remove bottlenecks in training, evaluation, batch inference, online inference, deployment … cost optimization. Strong systems engineering fundamentals across Linux, networking, storage, containers, Kubernetes, distributed runtimes, and production debugging. Experience with ML serving frameworks such as vLLM, Triton Inference Server, TensorRT, TorchServe, KServe, Ray Serve, or equivalent systems. Proficiency in Python for infrastructure automation, tooling, debugging, integration, and operational workflows. Practical understanding ...

Natural Language Processing (NLP) Researcher We tackle the most complex problems in quantitative finance, bringing scientific clarity to financial complexity from our London HQ. The role involves working with very large text corpora, applying the ...

distributed RL post-training systems, orchestrating trainer, rollout, environment, and reward workloads across thousands of GPUs. Build high-throughput rollout generation, integrating inference engines (vLLM, SGLang), weight synchronization, and asynchronous/off-policy schemes. Design RL environments for agentic, multi-step tasks — sandboxed code execution, tool use, computer use, multimodal … including sandboxed execution and multi-turn tool use. Deep familiarity with RL post-training frameworks (veRL, OpenRLHF, TRL, Ray orchestration) and rollout inference engines (vLLM, SGLang). Strong understanding of GPU clusters, networking, and communication libraries (NCCL, MPI) under mixed training and inference workloads. Nice to Have Running RL training ...