Lead architecture and implementation of LLM-first math library APIs and documentation, collaborate with stakeholders to deliver LLM-enhanced library releases, build metrics and tooling to measure LLM/agentic code-generation quality, and prototype solutions to make CUDA-X libraries LLM-friendly. Requires a MSc/PhD or equivalent experience, 3+ years experience, strong knowledge of LLMs (finetuning, RL, RAG, MCP), API design, and documentation.

Lead analytics and measurement work for Wikimedia's anti-abuse, safety, and security efforts. Responsibilities include designing experiments and metrics, performing quantitative research using Bayesian and frequentist approaches, collaborating with product, research, and engineering teams to develop and evaluate AI/ML models, building self-service dashboards, and working with large-scale data using tools such as Hive, Presto, Druid, Spark, Apache Iceberg, DBT, Airflow, and Superset. Fluency in Python or R, SQL, and experience with experimental design, statistics, and machine learning are emphasized.

Senior Solutions Architect role focused on large-scale distributed training and alignment of neural networks (including MoE models). Responsibilities include advising on NVIDIA's training stack (e.g., Nemotron), designing training/post-training infrastructure and workflows, optimizing GPU/HPC training efficiency, and collaborating with customers and NVIDIA product/research teams.

NVIDIA seeks a Senior Solutions Architect with 5+ years in neural network inference optimization to guide EMEA customers on large-scale, multi-node GPU inference deployments (MoE, quantization, speculative decoding, KV cache optimization). The role involves collaborating with NVIDIA product teams (Dynamo, TensorRT-LLM, NIXL), addressing interconnect and memory-bound challenges, and engaging the developer community through workshops and reference architectures. A Poland-specific salary range is provided.

Hands-on role guiding EMEA AI-native companies to train and deploy diffusion/image/video/multimodal models on NVIDIA accelerated stack. Responsibilities include architecture guidance (DiT, UNet, flow matching), single- and multi-GPU optimization, memory and attention bottleneck analysis, codec-aware video preprocessing and tokenization, and contributing technical feedback and demos. Requires MS/PhD or equivalent experience and 5+ years in AI/ML with strong computer vision and diffusion-model expertise. Poland base salary range explicitly listed: 292,500 PLN–507,000 PLN (annual).

Join the globally distributed Elasticsearch Search Relevance team to design and implement new search features, improve vector search capabilities, and resolve complex bugs. The role emphasizes Java development, work with vector similarity algorithms (e.g., HNSW, IVF), collaboration on open-source components like Lucene/Elasticsearch, and operating at scale in distributed systems.

Conduct research to advance intelligent systems, co-develop models and harnesses to improve memory and general intelligence in VLM/LLM systems, publish results, and open-source reproducible code. Requires a PhD or equivalent and at least 6 years of relevant experience in machine learning, AI, computational statistics, or applied mathematics, with proficiency in Python and a track record of research publications.

Lead development of decision-grade evaluation environments, benchmarks, and reproducible CI pipelines used to gate model releases and inform training and product decisions. Role requires 6+ years' experience with LLMs and large-scale evaluations, strong statistical foundations, and hands-on experience running workloads on GPU/HPC clusters (Slurm) and experiment management tools such as MLflow or W&B. Locations listed include Spain, Switzerland, Netherlands, France, and Poland (remote eligible).

DuckDuckGo is hiring a Staff Data Scientist to join the Data Science Functional Team. The role focuses on improving search ranking and relevancy through projects such as lexical matching, document-authority signals, search evaluations, and using LLMs for feature generation and ranking. The position requires 10+ years in Data Science (including 2+ years at Staff/Principal/Director level), production ML deployment experience, strong Python and advanced SQL skills, and experience with search infrastructures (e.g., Vespa, Elasticsearch/OpenSearch, Solr). The role is remote across many listed countries, expects ~40 hours/week on average, and requires travel ~2 times/year for company meetups. Compensation listed as $222,033 USD annually plus stock options.

Lead the Data Science functional team to set multi-team roadmaps, provide hands-on mentorship, and guide projects involving statistical modeling, experiment design, classifier implementation, ML/NLP and LLM work. Requires 10+ years in data science or ML engineering (including 4+ years leading teams), production-grade Python and SQL skills, familiarity with columnar databases (BigQuery, ClickHouse, Redshift, Druid), and experience designing experiments and privacy-aware analytics. Position is remote (locations listed), expects ~40 hours/week, and requires travel ~2x/year. Compensation listed at $243,800 USD annually plus stock options.

Senior Software Engineer for AI Inference Systems responsible for architecting and implementing high-performance inference stacks (e.g., vLLM), optimizing GPU kernels and compilers, defining benchmarking methodologies (MLPerf), and scaling workloads across multi-GPU, multi-node, and multi-cloud clusters. Requires strong skills in Python and C/C++, GPU programming (CUDA), performance profiling, containerization/orchestration, and significant experience in systems/performance engineering for ML.

NVIDIA seeks a Solutions Architect to work with customers and internal teams to evaluate, benchmark, and improve LLMs, VLMs, embedding models, and agentic pipelines. Responsibilities include driving adoption of NVIDIA evaluation tools (NeMo, Eval Harness, NVIDIA Evaluation Microservice), building proofs-of-concept, analyzing failures and mitigation strategies (model selection, retrieval optimization, fine-tuning), and collaborating with engineering, product, and sales. Requires MS/PhD or equivalent and 5+ years' hands-on experience building and evaluating deep neural networks; strong English communication skills.

Work on a performance-oriented, low-level inference infrastructure using hardware offloads (GPU kernels, CUDA) and RDMA network cards. Requirements include ~8 years in modern C++/C/Rust development, ~3 years in Linux development, deep knowledge of TCP/IP, computer architecture and OS concepts. Advantages include experience with LLM inference stacks, distributed storage, Linux internals, low-level optimizations, parallel programming, HPC and RDMA.

Senior role to develop scalable library software and optimize high-performance parallel numerical applications across CPU and GPU (CUDA). Requires 5+ years experience, strong C/C++ and software-development skills, performance tuning, benchmarking, and collaboration across distributed teams. Python and familiarity with CI/CD and profiling tools are desirable.