Tower Research Capital

In our ongoing AMA series, we're sharing answers to frequently asked questions on how we build, operate, and evolve one of the world’s most sophisticated trading platforms. Today's question: What lessons tend to come from rare but extreme market events, such as COVID or the Global Financial Crisis? A: Extreme market events tend to test assumptions that may not be visible in normal conditions. They can reveal where models, infrastructure, data pipelines, liquidity assumptions, or operational processes are more fragile than expected. For quants and engineers, one of the main lessons is the importance of building systems that are robust, observable, and adaptable. You may not be able to predict the next shock, but you can build tools that help teams understand what is happening quickly, identify whether an issue is coming from the market, the model, or the infrastructure, and respond in a disciplined way. These events also reinforce the value of humility. Historical data is essential, but it never captures every possible future regime. Strong research and engineering practices account for that by continuously monitoring live performance, questioning assumptions, and feeding lessons from production back into future development. Have more questions about our business, platform, or life at Tower? Let us know! Explore more questions and answers here: https://lnkd.in/gR7YNsgP

There's still time to visit us at [ICML] Int'l Conference on Machine Learning 2026! Stop by the Tower booth to meet the team, learn more about the work we do, and take on our Production Trading System Challenge. We're looking forward to connecting with fellow researchers, engineers, and builders during the final days of the conference.

ICYMI: AI in capital markets is moving from experimentation to execution. Earlier this year, Ramit, Tower's Global Head of Core AI & ML, joined industry leaders at Georgia Tech's AI and Future of Finance Conference to discuss what it takes to build production-ready AI systems. Key takeaways from the discussion: - Production-ready AI depends on more than powerful models. Success requires the right architecture, knowledge systems, and operational framework. - As AI agents become more capable, governance, security, observability, and cost management become increasingly important. - The most effective AI initiatives begin with clearly defined business objectives and a disciplined approach to where AI can create value. Read the full recap: https://lnkd.in/gR4dcbhg

Earlier this spring, Tower's Limestone trading team hosted its Fourth Annual Limestone Data Challenge at IIT Bombay. The competition brought together talented computer science students for a fast-paced day of quantitative problem-solving, data analysis, and development-focused challenges. Working in teams of two, participants applied their technical skills and creative thinking to tackle complex, data-driven problems inspired by the world of quantitative finance. The Limestone Data Challenge has become an important part of Tower's student engagement efforts, providing the next generation with hands-on exposure to the analytical and engineering challenges that power modern trading. Congratulations to all of this year's participants, and thank you to everyone who helped make the event a success. Read the full recap: https://lnkd.in/eQMKxsyc

Summer is in full swing at Tower, with teams across our global offices finding opportunities to connect, recharge, and enjoy time together beyond their day-to-day work. Here are a few highlights from June: Summer Celebrations  Teams across the globe came together to celebrate the season, while members of the Tower Women's Initiative Network (WIN) gathered for a summer social. Wellness and Recreation From yoga sessions in our India offices to pickleball in Hong Kong, employees found opportunities for movement, mindfulness, and friendly competition. Cultural Celebrations  Our Shanghai office marked the Dragon Boat Festival, taking time to celebrate an important tradition together. These moments help build the relationships and sense of community that connect our teams across offices and regions. Interested in life at Tower? Learn more: https://lnkd.in/gZTtn7mQ 

Tower is excited to sponsor and exhibit at [ICML] Int'l Conference on Machine Learning 2026, taking place July 6-11 in Seoul, South Korea. As one of the leading gatherings for the global machine learning community, ICML brings together researchers, engineers, and practitioners working at the forefront of AI and computational research. We're looking forward to engaging with attendees throughout the week, exchanging ideas on machine learning, research, and engineering, and learning from the broader AI community. If you're attending, stop by the Tower booth to meet the team and learn more about the technical challenges we work on. Learn more about ICML 2026: https://lnkd.in/g22S_HWf Explore careers at Tower: https://lnkd.in/gTXB9bBB

Earlier this month, Tower Research Ventures hosted a happy hour for AI founders and engineers as part of New York Tech Week 2026. Held at our rooftop, the event brought together founders, engineers, investors, and other members of New York's growing AI ecosystem for an afternoon of conversation, networking, and idea sharing. As AI continues to evolve rapidly, opportunities to connect with fellow builders and innovators have never been more important. Thank you to everyone who joined us!

Europe's rates markets are entering a new phase of competition. In a new article, Tower's Milain Thakkar explores the growing role of €STR futures, the challenges of building durable liquidity, and why long-term success in listed rates markets depends on more than simply launching a new contract. Key themes include: - The relationship between €STR and Euribor futures - Why liquidity, adoption, and market structure matter more than volume alone - The challenges and costs of fragmented liquidity - What venues must do to build lasting market participation As Milain writes, competition can create opportunity, but lasting success requires conviction from the market. Read the full article: https://lnkd.in/gvyNEVtv

Tower recently attended PyCon US 2026, where more than 2,000 software engineers, data scientists, researchers, and open-source contributors gathered to connect, learn, and share ideas. Throughout the conference, members of our trading, data, and HR teams spoke with approximately 1,000 attendees about Python's role in quantitative trading, software engineering, and large-scale data systems. We also hosted our inaugural Tower PyCon Challenge, with more than 120 participants putting their Python and problem-solving skills to the test. Read the full recap: https://lnkd.in/erdKGvHm

In our ongoing AMA series, we're sharing answers to frequently asked questions on how we build, operate, and evolve one of the world’s most sophisticated trading platforms. Today's question: How do you evaluate new languages, frameworks, or tools before adopting them at scale? A: Tower has an internal architecture group that plays a key role in evaluating new languages, frameworks, or tools before adopting them for critical parts of our trading stack. Proposals can be made by engineers across the organization. Depending on the complexity of the proposal, we might create a working group of relevant technology experts across the organization to help evaluate it. Our evaluation process often entails building a proof-of-concept to build confidence on the new technology and assess key stakeholder feedback, before committing to a large-scale adoption. Have more questions about our business, platform, or life at Tower? Let us know! Explore more questions and answers here: https://lnkd.in/gR7YNsgP