Chasing the Speed of Light - Software Engineer - Algorithmic Trading - AI - US - Elite Compensation Package

Chasing the Speed of Light - Software Engineer - Algorithmic Trading - AI - US - Elite Compensation Package By the time light crosses this room, our client has already lost ground. They are doing something about it. They need you to help. The Opportunity This is a firm widely regarded as setting the standard that others benchmark against. If you work in this industry, you already know who the handful of firms at that level are. This is not a role that surfaces often. The firm does not hire at volume. When a seat opens, it is because the business is growing into genuinely new technical territory, and they need someone capable of operating at the frontier of it, not someone to maintain what already exists. The business is built entirely on the quality of its technology and the people who build it. There is no product layer sitting above the engineering. The systems here are the edge. The Work The problems you would be working on sit at the intersection of two worlds that rarely overlap this cleanly: the sub-microsecond execution infrastructure that drives live trading, and the ML systems that are increasingly embedded directly within it. Most firms treat these as separate disciplines. This firm does not, and that distinction matters. On the low-latency side, the work spans order entry, market data ingestion, and pre-trade risk systems where performance is measured in nanoseconds and every architectural decision carries a real cost. Engineers here work close to the hardware: FPGA pipelines, kernel-bypass networking, custom memory management, and cache-aware data structures. The expectation is not that you know every tool on day one. The expectation is that when you encounter a bottleneck, you understand it completely before you touch it. On the systems side, the firm has invested heavily in ML infrastructure; training pipelines, inference systems, throughput optimisation, and the custom kernels that sit underneath all of it. If you have built systems where the gap between theoretical and actual performance is something you take personally, you will find problems here that are worthy of that instinct. Ownership is end-to-end. You design it, build it, ship it, and monitor it in production. The feedback loop between a decision you make and its real-world consequence is measured in hours, sometimes less. That is not a feature for everyone. For the right engineer, it is the most compelling thing about the job. The Team The engineering team is deliberately small. Headcount has never been a proxy for output here. The people you would be working alongside are, without exception, engineers who could have gone anywhere and chose this. Several arrived from the most selective firms in systematic trading and high-frequency execution. Others came from hyperscalers, where the work was excellent but the distance between their contribution and its impact had grown too large. The common thread is not a particular background or a particular stack. It is a specific instinct; an inability to leave a performance problem half-solved, and the technical depth to finish it properly. The team has no passengers. It also has no unnecessary hierarchy. Decisions are made by the people closest to the problem, and those people are trusted to make them well.The culture is rigorous but not rigid. People disagree openly and resolve those disagreements with evidence. There is a high tolerance for unconventional approaches and very little tolerance for sloppy thinking. If that sounds like the environment you have been trying to find, it probably is. Who we are looking for We are open to candidates from a range of backgrounds. What matters is depth, not title. If you have spent serious time working at the hardware boundary, whether in HFT, HPC, or high-performance systems research and you have the scars to prove it, we want to hear from you. Equally, if you have built ML infrastructure at scale and understand what it means to push a system to its theoretical limits, this firm is investing in exactly that space, and the timing is good. Strong C++ is important for most of the roles within this search. Python is used heavily on the research and ML infrastructure side. Familiarity with Linux internals, network stack optimisation, or GPU/FPGA programming will all be relevant depending on where you land within the team. We are less interested in whether you have used a particular tool than in whether you understand, at a fundamental level, why it works the way it does. Seniority is less important than substance. The firm promotes from within consistently and has the compensation structure to retain the people it wants. Engineers who join at a mid-level and prove themselves move quickly. We are equally interested in candidates who are already operating at a senior or staff level and are looking for problems that match their ability. CompensationCompensation is structured to reflect what the work demands which is to say it is among the best available anywhere in the industry. Base salary, performance bonus, and equity are all meaningful components of the package. Total compensation for the right candidate will be significant and will be discussed openly and early in the process. We do not believe in making good engineers run a gauntlet before finding out whether the numbers work. ProcessWe welcome approaches from engineers who are actively looking and from those who are open to a conversation but not formally on the market. Referrals are warmly encouraged; if you know someone who belongs in a room like this, we would be glad to hear about them. If you are the kind of engineer who finds the gap between theoretical and actual performance personally unacceptable, our client would very much like to meet you.