Why FIX Is More Important Than Ever

Trading, But Faster: How AI and Automation Are Transforming FIX and Trading Technology

Trading technology has evolved dramatically over the past two decades, but some of the processes sitting behind the world’s financial markets remain surprisingly manual. As artificial intelligence, automation and data become increasingly important across capital markets technology, firms are looking again at established workflows and asking a simple question: is there a better, faster and more efficient way to work?

In this episode of FinTech Focus TV, Harrington Starr CEO Toby Babb is joined by Sumit Kumar, Founder and CEO at Sosuv, to explore the evolution of FIX, the opportunities created by AI in trading technology and why accelerating traditionally manual processes can have a direct impact on revenue.

Drawing on around 20 years of experience in capital markets trading technology, Sumit discusses his journey from working directly with trading platforms and FIX connectivity to building a business focused on solving the challenges he had experienced first-hand. The conversation explores FIX certification, counterparty onboarding, trading data, artificial intelligence, automation and the future of financial technology, before turning to another critical ingredient behind technology businesses: attracting and retaining talented people.

FIX trading technology and the problem hiding behind the protocol

Sumit’s career has been centred on capital markets trading technology. During the first decade of his career, he worked primarily on trading platforms and FIX connectivity, including time at Goldman. That experience gave him a close view of the infrastructure, processes and challenges underpinning electronic trading.

It was also where he identified an opportunity.

FIX has established itself as a common messaging standard that enables different participants across the financial markets to communicate. But while the protocol itself has evolved and fulfilled an important role within the industry, Sumit argues that many of the processes surrounding it have not evolved at the same speed.

FIX certification, support and onboarding can remain manual, fragmented and heavily dependent on engineering resources. Rather than simply accepting these processes as an unavoidable part of trading technology, Sumit and his team began asking whether they could be redesigned.

That first-hand experience became the foundation of Sosuv. The company initially worked on engineering services and projects involving areas such as platform migration and FIX connectivity. Through those projects, the team developed significant domain knowledge while repeatedly encountering the same pain points within trading technology workflows.

Eventually, the business began turning that accumulated expertise into technology of its own.

The result was SoFIX, a product designed to automate the pain points the team had encountered over years of working on these projects. Sumit explains that the platform can automate the certification process end to end, using AI alongside other technologies. Processes that previously took weeks, and in some cases months, can potentially be reduced to hours or days.

That transformation sits at the heart of the episode: taking an established part of capital markets technology and asking how automation can make it work significantly faster.

Trading automation can turn manual processes into software processes

The opportunity Sumit describes is not simply about adding AI to an existing workflow. It is about fundamentally reconsidering how that workflow operates.

He explains how manual tasks surrounding FIX certification have traditionally required engineers to repeatedly work through processes that can become a case of “rinse, repeat”. Sosuv’s approach is to turn more of that work into a software-driven process.

One example discussed in the episode is the handling of a 200-page FIX specification. Traditionally, a person might need to read through the specification before working through configuration files and the subsequent certification process. Technology can instead be used to ingest that specification, generate configuration files, create test cases, execute those tests and validate the results.

The goal is end-to-end automation of a certification lifecycle that could previously consume days, weeks or even months.

For financial institutions, this represents more than a technical improvement. It demonstrates why trading technology innovation often matters most when it addresses a tangible business problem.

Toby highlights that the combination of efficiency, productivity, speed, control and revenue makes this very different from solving a technical issue simply because the technology exists. It addresses a process that market participants have experienced for years and often accepted as part of doing business.

For firms investing in electronic trading technology, FIX connectivity and capital markets infrastructure, challenging those assumptions could unlock considerable value.

FIX onboarding is a revenue problem, not just a technology problem

One of the most significant ideas to emerge from the conversation is Sumit’s argument that slow FIX onboarding should be viewed through a commercial lens.

When firms onboard a new counterparty, he explains, the measurement is often binary: has that counterparty been onboarded or not?

What is not always measured closely enough is how long the process took.

If an onboarding takes two months, could it have been completed in two weeks? Could it have been completed in two days?

This changes the nature of the conversation. Rather than viewing slow onboarding solely as a technology or process challenge, Sumit describes it as a P&L and revenue problem. The longer it takes to complete onboarding or certification, the longer it takes for the associated revenue to begin coming into the organisation.

For trading technology businesses and financial institutions, speed therefore has a direct commercial implication.

Sumit says large institutions using the platform have reported around 80% efficiency in effort and processes becoming four or five times faster. The value of automation, from this perspective, is not automation for its own sake. It is the ability to reduce effort, accelerate processes and ultimately bring revenue into an organisation sooner.

As financial services firms continue to invest in trading infrastructure, this connection between technology performance and business outcomes is likely to remain critical. Technology teams are not operating in isolation from the commercial goals of an organisation. The systems, processes and platforms they build and maintain can directly influence how efficiently that organisation serves clients and generates revenue.

AI in trading technology needs to solve a genuine problem

Artificial intelligence inevitably enters the conversation, but Sumit’s approach to AI is deliberately measured.

As Toby notes, almost every technology business now has an AI story. Traditional SaaS businesses are increasingly describing themselves through the lens of artificial intelligence, while the wider financial technology industry continues to explore how generative AI and automation could reshape existing operations.

For Sosuv, however, the starting point was not AI.

It was the problem.

Sumit explains that the company’s strength lies in its domain expertise. Its team experienced the pain points within trading technology workflows directly, understood traditional technology and programming, and then layered AI into that existing knowledge.

This means using artificial intelligence where it creates real value rather than introducing it simply because AI is fashionable.

That distinction is particularly important within financial services and trading technology, where governance, security and reliability cannot be afterthoughts. Sumit argues that AI cannot simply be implemented everywhere and left to operate without oversight. The human in the loop remains important.

For example, AI can provide considerable value when parsing a 200-page FIX pack. But where a problem can be solved effectively through traditional programming, Sosuv continues to use traditional programming.

The result is a combination of established engineering practices and emerging artificial intelligence capabilities rather than a wholesale replacement of one with the other.

AI in financial services still needs governance and human oversight

Working with large financial institutions places additional demands on technology providers.

As Sumit explains, a product cannot simply look impressive in a demonstration. For financial institutions to adopt technology, it must sit within an appropriate framework of governance, compliance and security.

This is also reflected in how Sosuv approaches AI internally.

The company encourages its developers to use artificial intelligence where it can help them become faster. Engineers who previously wrote code without these tools can now use AI-assisted development to accelerate parts of the process.

But that does not mean allowing AI-generated code to move directly into production without scrutiny.

Human review, governance and guardrails remain necessary. For organisations building financial technology products, this combination could become increasingly important as AI adoption grows. The opportunity is not necessarily to remove people from technology workflows, but to determine where AI can improve productivity while maintaining the controls required by financial institutions.

That has implications for financial technology talent too. As AI becomes embedded into software engineering and trading technology roles, professionals will increasingly need to understand how to use these tools effectively while retaining the domain knowledge, engineering judgement and governance awareness required in highly regulated environments.

For a FinTech recruitment business such as Harrington Starr, these developments are important because changes in technology inevitably influence the skills businesses need from their people.

The future of FIX could be shaped by AI and trading data

FIX has been part of the financial markets for decades, but Sumit does not see its relevance diminishing. In fact, he believes advances in technology could make FIX more useful.

A major reason is data.

Sumit argues that the information contained within FIX logs has historically been underappreciated. Much of the industry’s attention and investment has naturally focused on the front-end experience and the trading platforms visible to users, while FIX infrastructure has often remained in the background, associated with support and onboarding teams.

AI could change what organisations are able to extract from that information.

FIX logs contain trade-routing data, creating opportunities to analyse far more than simply whether something failed. Sumit describes a progression from identifying what failed, to understanding why it failed, to determining what needs to change in preparation for the future.

That introduces the possibility of becoming more predictive through the application of AI to FIX data.

Rather than seeing FIX as a historical standard at risk of losing relevance as financial technology advances, the discussion suggests the opposite. Emerging technology could provide new ways of extracting intelligence from infrastructure and data that the industry has possessed for years.

Sumit also points to adoption in areas including crypto, post-trade workflows and market data. As backend processes become more efficient, the role of FIX could continue to evolve alongside the wider trading technology ecosystem.

Trading data will become increasingly valuable in 2027 and beyond

Looking towards the remainder of 2026 and into the coming years, Sumit expects the importance of data to continue growing.

AI and automation depend heavily on the quality of the information available to them. As a result, organisations will need to place greater focus on ensuring their data is clean and useful.

Sumit expects increased investment in data quality and says he is already seeing institutions look more closely at FIX logs and other application logs to identify the intelligence that can be extracted from them.

At the same time, he expects AI adoption to move beyond the chatbot.

Much of the early enterprise experience of generative AI has centred on conversational interfaces. The next stage could involve AI becoming more deeply embedded into trading workflows, working alongside traditional technology to automate processes and create efficiencies.

AI adoption and automation, in Sumit’s view, will increasingly go hand in hand.

This makes data impossible to ignore. Wherever valuable data sits within an organisation, the ability to structure, understand and use it could increasingly determine how successfully businesses deploy AI and automation.

For technology leaders, trading firms and financial institutions, the challenge is therefore not simply deciding which AI tools to adopt. It is ensuring the underlying technology, data and talent foundations are strong enough to make those investments valuable.

FinTech recruitment and the importance of attracting technology talent

Technology alone does not build a successful FinTech business, and the latter part of the conversation turns towards the people behind Sosuv’s growth.

Sumit describes himself as fortunate to have a strong team of high performers around him. The company operates with a distributed structure, with business-focused employees in key commercial locations and its engineering team based in Bangalore.

He credits the team and its leaders with helping the company reach its current position, alongside the trust built with clients. Sosuv has worked with major financial institutions, with some relationships extending for more than a decade, and Sumit highlights work ethic, service quality and trust as important parts of that success.

As the customer base and team continue to grow, the company is also considering further expansion into locations including New York, London and potentially Hong Kong.

This leads Toby and Sumit into a wider discussion about attracting and retaining the best technology talent.

For businesses competing for software engineers, trading technology professionals and specialist financial technology talent, Sumit’s view is straightforward: hire smart people and make sure they continue to be challenged.

FinTech technology talent needs challenge, learning and growth

Hiring talented people is only the beginning. Retaining them requires creating an environment in which they can continue developing.

Sumit argues that smart people need to be challenged. Employees need to understand the company’s vision, but they also need to see what their own growth looks like within that journey.

Is the company growing? Are they growing with it? What are they learning?

When people are aligned with the vision, understand the product roadmap and growth story, and are given challenging, world-class work, Sumit believes they are more likely to stay.

His central message is that retaining top talent requires providing both growth and learning.

This resonates strongly with the challenges Harrington Starr sees across FinTech recruitment and financial technology hiring. As artificial intelligence, automation, data engineering and trading technology continue to evolve, organisations need people capable of evolving alongside them. The strongest technologists are often motivated not only by the technology they use today, but by the problems they will have the opportunity to solve tomorrow.

Toby reinforces this point, arguing that the best technologists want to grow and develop. Businesses that recognise this and deliberately design environments around challenge, opportunity and progression put themselves in a stronger position to attract the people capable of driving innovation.

The future of trading technology is faster, smarter and more automated

This episode of FinTech Focus TV demonstrates how innovation in financial technology does not always require replacing the infrastructure that came before it. Sometimes the biggest opportunity lies in taking an established process, identifying where friction remains and applying new technology intelligently.

For Sumit Kumar and Sosuv, that process began with years of first-hand experience in FIX connectivity and trading technology. The team encountered manual, fragmented and engineering-intensive workflows, developed deep domain knowledge around those challenges and then began turning that expertise into a product designed to automate them.

The wider lessons extend beyond FIX certification.

AI has enormous potential across financial services, but its value comes from solving genuine problems. Automation can make processes faster, but that speed becomes far more meaningful when connected to commercial outcomes. Data has existed throughout financial markets infrastructure for decades, but emerging technologies could allow firms to derive far greater intelligence from it. And none of these developments remove the importance of skilled people who understand the technology, the domain and the problems businesses are trying to solve.

As trading technology continues to develop, the intersection of AI, automation, data and human expertise will become increasingly important. Firms will need strong technology infrastructure, robust governance and high-quality data, but they will also need software engineers, FIX specialists, data professionals, trading technology experts and technology leaders capable of turning those tools into meaningful outcomes.

For Harrington Starr, as a specialist FinTech recruitment business working across the global financial technology market, conversations like this highlight how rapidly the skills landscape is changing. The future of financial technology recruitment will increasingly be shaped by businesses seeking professionals who can combine deep capital markets knowledge with modern engineering, data and AI capabilities.