Q: What does TigerGraph do?

TigerGraph is an enterprise-grade AI infrastructure and graph analytics platform. We help organizations make sense of complex, interconnected data. Think of it as revealing the "what behind the why" by exposing the relationships within data. 

Unlike traditional queries that return isolated results, TigerGraph provides explainable answers – showing how different data points connect and influence each other. This is crucial for situations with multi-hop queries and deep, intricate relationships. 

Many of our customers build their AI applications on top of our data infrastructure and many AI infrastructure capabilities are built on TigerGraph. We occupy a crucial position in the AI tech stack, bridging the gap between large language models (LLMs) and end-user applications, helping customers to maximize the potential of LLMs without building them directly or creating consumer-facing apps. 

‍

Q: What types of problems does TigerGraph solve for customers?

A: We primarily serve large enterprises, including major financial institutions like JP Morgan, Bank of America, and Mastercard, and technology companies like Microsoft by offering flexible deployment options for a cloud-native offering, a native platform for on-premise, and hybrid solutions.

The pain points we address vary by industry. For large banks, fraud detection and anti-money laundering are primary use cases. We provide unique visibility into multi-level sourcing, quickly identifying suspicious patterns.

Another key application is creating virtual supply chains. Large organizations often rely on thousands of microservices. TigerGraph helps technology teams troubleshoot this complex infrastructure by building digital twins – providing a clear picture of how components interact.

At our core, we excel at entity resolution. This means making sense of master data and complex relationships, identifying individual elements even with inconsistent representations (e.g., a name spelled multiple ways). We resolve these discrepancies, establishing a single, unified understanding.

This extends to cybersecurity, where we help detect suspicious network activity and provide explainable insights for improved threat detection and supply chain visibility.

‍

Q: What is differentiated about your approach?

A: Our core differentiator is our pioneering native parallel graph database architecture. It's a truly distributed design enabling massively parallel processing, unlike single-machine databases. This allows TigerGraph to handle both real-time transactions (OLTP) and complex analytical queries (OLAP) with high performance – a critical need that traditional databases, typically optimized for one or the other, struggle to meet.

Our architecture also handles varying data sizes efficiently, from gigabytes to petabytes. Over 300 native data connectors unify information from diverse sources, creating a holistic view.

This unified data is essential for modern AI. AI applications require not just vast amounts of data, but also contextual understanding. TigerGraph's engine delivers insights with clear data lineage – providing the explainability we need to actually trust AI-driven decisions. We help customers move beyond the "black box" model, offering transparency and traceability.

‍

Q: TigerGraph has been an early pioneer in the AI space–how has the recent acceleration in the space impacted the business?

A: The AI surge has been very positive for us as a company. Our enhanced vector attribute support is driving strong customer interest in hybrid search capabilities – a key area for building AI applications. We expect this to accelerate innovation and growth.

A major breakthrough is our collaboration with NVIDIA on GPU-accelerated Graph Neural Network (GNN) training. GNN models, traditionally resource-intensive, can now be trained 100 to 200 times faster on NVIDIA GPUs. This is solidified by NVIDIA's investment in TigerGraph. A major financial institution is our flagship customer for this technology.

This significantly strengthens our market position. Companies investing in GPUs want maximum ROI and efficient model training, and that speed improvement is a game-changer. We're planning a joint announcement.

‍

“Unified data is essential for modern AI applications, which require contextual understanding. TigerGraph's engine provides the explainability we need to actually trust AI-driven decisions.”

‍

Q: Are there any upcoming milestones or exciting things happening right now?

A: We recently launched TigerGraph Savanna, our cloud-native offering, with true storage-compute separation. This allows customers to scale efficiently and optimize cloud infrastructure while maintaining our performance. Savanna is currently on AWS, with plans to expand to other major cloud platforms.

We're also enhancing the platform with integrated vector database support – a highly anticipated feature complementing our graph capabilities. This will roll out to all customerssoon.

Beyond that, we're deepening collaborations with major cloud providers (AWS, GCP, Azure), working with their key customers. We're expanding support for deployments in private cloud environments and dedicated storage buckets, ensuring seamless integration with existing infrastructure.

‍

Q: How do you think about the long-term vision for the company?

A: Our long-term vision is to capitalize on the growing cloud database market and the demand for graph solutions. We see huge potential in empowering customers to build AI applications on our platform, especially as explainable AI and data lineage become crucial.

We're focused on expanding our enterprise customer base, particularly cloud-native companies, and strengthening our presence in key regions like the Americas and Western Europe. We aim to become the foundational technology upon which customers build diverse applications.

We envision customers using TigerGraph for multiple use cases beyond initial deployments. For example, a customer using us for fraud prevention could easily extend to build digital twin infrastructures or create customer 360 views. This expansion into adjacent use cases, combined with regional growth and new AI applications, will drive our future growth. 

‍

Q: What has your background been up until joining TigerGraph?

A: Joining TigerGraph six months ago was a natural progression after my time at Google, where I worked on AI/ML platforms and customer engagement. Having witnessed the rise of LLMs firsthand and understanding the foundational importance of Knowledge Graphs, I was drawn to TigerGraph's mission of empowering customers to build similar solutions. 

My past experience as an entrepreneur and VC, combined with the company's world-class client list drew me to join the team. Six months later, I'm more excited than ever about the opportunities and customer impact that are in store.

‍

Q: What advice would you give to other founders/tech leaders?

A: Focus is paramount, especially for platform companies. It's less about pursuing every opportunity and more about excelling in a few key areas. Focusing our efforts allows us to deliver exceptional value to our target segments and ensure we don't spread limited resources too thin. Success is more often prioritizing what you should do, rather than what you could do.

‍

Q: How can an investor best partner with and support an early-stage company like TigerGraph?

A: The executive team and investor partnership is critical. Startup journeys are hard, and executives and investors must establish a relationship of trust. This trust isn't blind faith; it's about unwavering support and constructive challenge. When things thrive, we push for greater ambition. When challenges arise, we collaborate, offer ideas, and help identify blind spots. This requires active, open dialogue, not top-down directives or isolation. The healthy investor-startup relationships I’ve experienced balance accountability and flexibility. 

‍

Q: What are you reading?

A: I enjoy reading fiction when I have free time. I find spy thrillers and mythological stories are a great way to provide a much-needed mental recharge.

‍