Electronic market maker trading derivatives, equities, ETFs, bonds, and FX globally
Optiver is a privately held market maker operating across nine global offices, providing liquidity to exchanges using proprietary capital. The tech stack reveals a quantitative powerhouse: heavy Python/C++/R foundation overlaid with GPU-accelerated ML (PyTorch, JAX, TensorFlow, CUDA, cuDNN) and low-latency infrastructure (FPGA, VHDL, Verilog, DPDK, Ethernet switching). Engineering and research dominate the 261-person hiring pipeline, with active projects spanning stochastic derivative modeling, high-frequency trading strategy analysis, and trading system automation—signaling a shift from traditional market-making toward AI-driven pricing and execution.
Optiver is a global electronic market maker founded in 1986 and headquartered in Amsterdam. The firm trades listed derivatives, cash equities, ETFs, bonds, and foreign currencies, committing capital at its own risk to provide liquidity across multiple exchanges. Operations span nine offices: Amsterdam, Sydney, Chicago, Austin, Taipei, Hong Kong, Shanghai, London, and Singapore. The company employs 1,001–5,000 people, with active hiring concentrated in engineering (103 open roles), research (30), and finance (29). Internal challenges center on compute platform stability, network infrastructure performance, trading system optimization, and scaling ETL pipelines to handle high-quality data at scale.
Python, C++, C, R, Java, and C# form the core. For low-latency and hardware work: VHDL, Verilog, SystemVerilog, and FPGA. GPU compute relies on CUDA, cuDNN, cuBLAS.
PyTorch, JAX, TensorFlow, XLA, Flax, Triton, and Keras. GPU libraries include CUDA, cuDNN, and cuBLAS. Inference and training pipelines are a major active project.
Eight countries: Australia, United States, Netherlands, United Kingdom, China, India, Singapore, and Taiwan. Engineering roles account for 103 of 261 active positions.
Other companies in the same industry, closest in size