When Hupo co-founder and CEO Justin Kim first started building the company, its mission looked very different from what it is today. Originally launched around four years ago under the name Ami, the startup focused on mental wellness and behavioral change—helping people manage stress, build habits, and improve emotional resilience.

At the time, the idea centered on how individuals respond to pressure and how long-term behavior is shaped. But over time, that direction evolved significantly. Today, Hupo operates in a completely different space: AI-powered sales coaching for enterprise financial services teams.

The transformation reflects a broader shift in how AI startups are rethinking product-market fit—not just building tools around abstract personal improvement, but embedding intelligence directly into high-value professional workflows.

From Human Behavior to Performance Systems

Kim has described his long-standing interest in performance, shaped in part by his passion for competitive sports such as basketball, football, Formula 1, and MMA. That interest eventually influenced how he thought about work performance in corporate environments.

Rather than focusing solely on wellness or motivation, he began to study how performance varies across teams in high-pressure environments. One recurring insight stood out: results often depend less on individual motivation and more on training quality, feedback consistency, and confidence under pressure.

That realization became the foundation for Hupo’s pivot.

Instead of building a consumer-facing mental wellness product, the company began reorienting itself toward enterprise performance systems—specifically targeting industries where communication quality and decision-making directly impact revenue.

Why Sales Coaching Became the Focus

Hupo’s current product is centered on AI-driven sales coaching, particularly for organizations in banking, insurance, and financial services (BFSI).

In these industries, sales performance is notoriously inconsistent across teams. Large organizations often struggle with scaling training effectively, especially when managers cannot directly monitor every customer interaction or provide real-time feedback.

Hupo’s approach aims to address that gap by using AI systems that can analyze conversations and provide immediate coaching insights during or after client interactions. Instead of relying solely on periodic training sessions, the platform integrates feedback into daily workflows.

According to Kim, the shift from wellness to sales coaching was not a complete break from the company’s original vision. In both cases, the underlying problem is human performance at scale—how individuals improve when systems, feedback, and context are limited.

Backing From Meta and Enterprise Expansion

Hupo’s early development received backing from Meta, which participated in its seed round. That early support helped the team refine its product direction, particularly around a key insight: software only works when it fits naturally into existing human behavior rather than forcing users to adopt entirely new routines.

That lesson carried through the company’s pivot and now shapes how Hupo designs its enterprise tools.

The startup has since raised a total of $15 million in funding, including a $10 million Series A round led by DST Global Partners, with participation from investors such as Collaborative Fund, Goodwater Capital, January Capital, and Strong Ventures.

Enterprise Customers Across Global Financial Markets

Hupo has already gained traction across multiple regions, particularly in Asia-Pacific and Europe. Its customer base includes major financial institutions and corporations such as:

• Prudential
• AXA
• Manulife
• HSBC
• Bank of Ireland
• Grab

These organizations operate in highly regulated environments, where sales processes are complex, compliance-heavy, and difficult to standardize across large teams.

Despite these challenges, Hupo reports strong expansion within its client base, with contracts reportedly growing significantly within the first months of adoption. This suggests that once embedded into enterprise workflows, the system delivers enough value to justify broader deployment.

Why BFSI Is a Difficult but Valuable Market

The banking and insurance sector is known for being one of the most challenging environments for early-stage tech companies. Strict regulations, long sales cycles, and conservative procurement processes often make adoption difficult.

However, it is also a sector where small improvements in sales performance can translate into substantial financial gains.

Hupo positions itself directly in this gap—offering tools that help scale coaching across thousands of employees without requiring constant managerial oversight.

Kim argues that traditional coaching systems simply cannot keep up with the scale and complexity of modern financial institutions. AI, in contrast, can analyze large volumes of conversational data and deliver feedback consistently across entire organizations.

The Technology Behind Real-Time Coaching

A key component of Hupo’s platform is its ability to process and understand conversations in real time. The system is designed to operate in regulated financial contexts, where accuracy, compliance, and tone are critical.

Unlike generic AI tools, Hupo’s models are trained specifically on:

• Financial products and services
• Common customer objections
• Client segmentation patterns
• Regulatory requirements and constraints

This domain-specific training allows the system to generate more relevant coaching insights tailored to real-world financial sales interactions.

Kim has emphasized that many AI coaching tools fail because they are built as general-purpose systems first and adapted later. Hupo instead started with the industry itself, designing the technology around how banks and insurers actually operate.

Leadership Experience Shaping Product Direction

Kim’s background has played a significant role in shaping the company’s direction.

He began his career at Bloomberg, where he worked in enterprise software sales targeting banks, asset managers, and insurers. That experience gave him firsthand exposure to the complexity of regulated financial systems and enterprise buying behavior.

He later worked at Viva Republica, the South Korean fintech company behind the Toss platform, where he was involved in product development focused on real user behavior and digital financial services.

These experiences helped him understand three critical perspectives simultaneously:

• How enterprise buyers evaluate software
• How end users interact with financial tools
• How regulated industries constrain product design

Hupo, in his view, sits at the intersection of these three realities.

Expansion Plans and Future Direction

The company is now preparing to expand into the United States, where distribution-heavy financial models create additional demand for scalable coaching systems.

The next phase of development includes:

• Real-time coaching enhancements
• Expanded enterprise deployments
• Growth in banking and insurance markets
• Scaling go-to-market operations globally

The broader ambition extends beyond sales coaching. Over time, Kim envisions Hupo evolving into a system that helps large organizations improve performance at scale across tens of thousands of employees—not just in sales, but potentially across broader operational roles.

Conclusion: A Pivot That Reflects a Larger AI Trend

Hupo’s transformation from a mental wellness startup into an AI-powered enterprise coaching platform reflects a wider trend in the AI industry: startups are increasingly abandoning broad, abstract consumer ideas in favor of highly specific enterprise use cases.

By focusing on financial services sales performance, Hupo is targeting a space where AI can deliver measurable, immediate impact—especially in environments where consistency, training, and communication quality directly influence revenue outcomes.

Whether the company expands beyond sales coaching or deepens its hold in BFSI will depend on how effectively it can scale its real-time AI systems across some of the world’s most complex and regulated industries.

Video games are often criticized for harming attention, encouraging addiction, or “rotting the brain,” but a growing body of psychological research is exploring a more nuanced possibility: certain types of games may actually help protect mental health under specific conditions.

One of the most frequently studied examples is the classic puzzle game Tetris, which has become an unexpected focus in experiments related to trauma processing, memory formation, and even addiction cravings.

While the evidence is still preliminary and based on relatively small-scale studies, the findings suggest that visually demanding games may have measurable effects on how the brain processes emotional experiences.

How a Simple Puzzle Game Became a Mental Health Experiment

Tetris works by requiring players to rotate and position falling geometric shapes so they fit together without gaps. On the surface, it is a simple arcade-style game. But cognitively, it heavily engages the brain’s visual and spatial processing systems.

This characteristic is what drew researchers to it in the first place.

A key hypothesis in psychological research is that traumatic experiences are stored in memory partly through vivid sensory and visual impressions. These memories can later resurface as intrusive flashbacks, a hallmark symptom of conditions such as post-traumatic stress disorder (PTSD), depression linked to trauma, and complicated grief.

Researchers began exploring whether occupying the brain’s visual-spatial systems shortly after trauma exposure might interfere with how these memories are consolidated.

Early Research: Interrupting Traumatic Memory Formation

One of the most widely cited lines of research in this area was led by clinical psychologist Emily Holmes at the Karolinska Institute.

Early experiments suggested that there may be a short window of time after a traumatic event during which memories are still stabilizing. During this period, visual memories are not fully fixed and may still be influenced by competing cognitive activity.

In controlled studies, participants who were shown distressing images or footage were later asked to play Tetris shortly afterward. Researchers observed that those who played the game experienced fewer intrusive visual memories in the following days compared to control groups.

The proposed explanation was not that the game “erases” memory, but rather that it competes for the same mental resources needed to consolidate visual aspects of traumatic experiences.

Testing the Idea in Real Trauma Patients

More recent research moved beyond laboratory simulations and into real-world clinical settings.

In one study, researchers recruited patients who had recently experienced actual traumatic events, such as traffic accidents, and arrived at hospital emergency departments within hours of the incident.

Participants were divided into two groups:

• One group played approximately 20 minutes of Tetris on a handheld console shortly after admission
• A control group engaged in other hospital activities such as reading, texting, or doing puzzles

When researchers followed up one week later, they found a noticeable difference in reported flashbacks.

The Tetris group experienced significantly fewer intrusive memories compared to the control group, with reductions reported at more than half in some measurements. However, longer-term follow-ups (such as one month later) did not show strong or consistent differences in overall mental health outcomes between the groups.

This suggests a more limited conclusion: the intervention may reduce early intrusive symptoms, but does not yet demonstrate long-term therapeutic effects.

Why Visual-Spatial Games Might Affect Cravings and Addiction

The same cognitive mechanism has also been explored in relation to addiction and craving behavior.

Cravings for substances like nicotine, alcohol, or even food are often accompanied by vivid mental imagery. People may “see” themselves engaging in the behavior, which can intensify desire and make resistance harder.

Researchers have tested whether engaging the brain in a visually demanding task—such as Tetris—can reduce the intensity of these cravings.

In a small experimental study, participants who played short sessions of Tetris after reporting cravings showed a measurable reduction in craving strength compared to those who did not play. The reduction was not dramatic, but it was consistent enough for researchers to consider it meaningful in behavioral terms.

The proposed explanation mirrors the trauma research: the game occupies the same visual cognitive resources that otherwise fuel internal imagery of the desired behavior.

Not a Cure, but a Low-Risk Intervention

Across studies, researchers consistently emphasize caution. Sample sizes are small, methodologies vary, and replication is still needed before firm clinical conclusions can be made.

However, one point stands out across multiple papers: the intervention appears to be low risk. Unlike pharmacological treatments or intensive therapy programs, short sessions of a simple game like Tetris do not introduce meaningful side effects.

Because of that, researchers have described it as a “low-intensity psychological intervention” rather than a formal treatment.

It is also important to note that the effect is not limited to Tetris specifically. Any task that strongly engages visual-spatial processing—such as certain puzzle games, drawing activities, or pattern-based games like Candy Crush—may produce similar short-term cognitive competition.

What This Research Actually Means (and What It Doesn’t)

These findings do not mean video games are a treatment for trauma or addiction. Nor do they suggest that gaming prevents mental illness in a general sense.

What the research does suggest is more specific:

• The brain has limited cognitive bandwidth for visual imagery
• Engaging that system can temporarily reduce intrusive mental images
• This may influence short-term symptoms like flashbacks or cravings

But it does not yet demonstrate long-term prevention or recovery from mental health disorders.

At best, this line of research points to a possible supportive tool that could one day complement established therapies—not replace them.

A More Nuanced View of Games and the Brain

The broader implication is that video games are not inherently harmful or beneficial. Their effects depend heavily on design, context, timing, and psychological state.

In this case, a simple puzzle game becomes interesting not because it is “therapeutic” in itself, but because it reveals something about how memory and attention compete in the brain.

Even if the clinical applications remain uncertain, the underlying idea is compelling: sometimes, what the mind focuses on in the immediate aftermath of intense experience can subtly shape how those experiences are later remembered.

And that, researchers argue, is worth continued investigation.