Why We Invested in Upsonic

Meryem Doğan
Oct 14
5 min read

The Rise of the AI Agent Era

In every period of transformation, some teams only follow the trend, while others set the standard. We believe that Upsonic belongs to the second category in this transformation.

Over the past few years, generative AI has initiated a brand-new era in human–machine interaction. However, the true power of this transformation now lies not only in the models but in the AI agent ecosystem that is taking shape around them.

According to 2030 forecasts of the global AI market, the AI agents market will be valued at USD 7.84 billion in 2025 and is expected to reach USD 52.62 billion by 2030, growing at a CAGR of 46.3%.¹ In addition, the “global enterprise agentic AI” market size is expected to be worth around USD 171 billion by 2034, from USD 5.3 billion in 2025, growing at a CAGR of 47.2% during the forecast period from 2025 to 2034.²

In particular, the infrastructure layer defined as the “AI Agent Framework” is the fastest-growing segment of the market. Large technology companies such as Google, Microsoft, and Anthropic are now investing more in agent-framework ecosystems rather than model development.

Upsonic focuses on the “AI agent creation and orchestration” layer within this market, serving both developers and enterprise customers. Companies are now transforming from structures that “use” models to ones that work with agents. From banking to the legal sector, from manufacturing to insurance, agent-based automation systems are redefining the workforce.

The fintech sector is one of the most concrete examples of this transformation. In financial technologies, where operational processes are still heavily dependent on humans, Upsonic eliminates this dependency by turning repetitive operations into agentic workflows. Processes such as KYC (Know Your Customer), risk analysis, compliance reporting, and fraud detection can now run securely and transparently on Upsonic’s AgentOS infrastructure. This creates a significant advantage for fintech companies in terms of both operational speed and reliability.

Upsonic stands at the very center of this transformation, as an open-source platform that enables developers, enterprises, and startups to easily build, manage, and scale their own AI agents.

The Problem: Building AI Agents Is Still Hard

Building an AI agent is still a challenging process: Layers such as LLM integration, memory management, knowledge base, tool integrations, and hallucination-prevention algorithms require complex engineering.

Building reliable, low-cost yet scalable systems that can operate safely in production environments is still technically beyond the reach of many organizations.

This is exactly where Upsonic comes in. It offers an AI Agent Framework that allows users to create and run their own agents in just a few steps, without the need to build infrastructure from scratch.

This framework can work in integration with different model providers such as OpenAI, Anthropic, and DeepSeek; it also provides memory, knowledge base, task management, and reliability modules in a single package.

The Technology: Reliability-First, Developer-Centric

Upsonic’s technical vision is not just about “creating agents,” but about delivering production-ready agents.

Reliability Layer: Develops special control mechanisms to address one of the biggest problems of LLMs—hallucination.
Multi-Model & Multi-Tool Integration: Among the first frameworks to work with Anthropic’s MCP (Multi-Tool Agent) system.
Open Source + API-First: Combines community-driven development with strong enterprise compliance.
Dynamic Agent Quality Testing: Features an automated quality monitoring system that measures agent performance and enables self-healing structures.

Upsonic’s differentiation lies not only in technical capability but also in its reliability-first design and multi-model architecture, which make it one of the rare frameworks that allows developers to move into production environments with confidence.

This focus on reliability is particularly critical in sectors such as fintech, where regulation and data security are paramount. Upsonic’s AgentOS architecture enables financial institutions to run operational agents in an auditable and highly secure manner. This approach paves the way for large-scale adoption of agent-based automation in the fintech ecosystem.

The Market: Where Software Becomes Autonomous

Software is no longer just automated but becoming autonomous. AI agents have become systems that can make their own decisions without human intervention, connecting the data and action layers. The AI agent framework segment in which Upsonic operates offers enormous potential in terms of both enterprise automation and developer productivity. According to MarketsandMarkets and Grand View Research, this field will reach a market size of USD 50–52 billion by 2030 and grow at an annual average of more than 45%.¹³

Recent major developments in the industry also confirm this direction:

Microsoft introduced its infrastructure called the “Agent Framework,” making agent development cloud-based.
Google aims to increase the interoperability of multi-agent systems with the “Agent Development Kit (ADK).”
Anthropic is accelerating standardization at the framework layer with the “Model Context Protocol (MCP).”

In addition, new protection protocols are being developed against “agent hijacking” and “prompt injection” attacks in the fields of security and resilience.⁴⁵

In this environment, Upsonic serves as a modern bridge, open source but enterprise-ready, allowing developers to quickly implement their agent infrastructures.

The Team: Resilience, Vision, and Execution

The story of Upsonic is actually a story of a pivot. Our first meeting with the founding team was at an event when they were still working on a different data product. Over time, their ability to quickly adapt to changing market conditions while reshaping their product strategy without losing their vision was one of the first elements that impressed us.

Mümtaz Vural, Onur Ulusoy, and Mehmet Bartu are people who have been working together for a long time, with deep technical expertise and a strong entrepreneurial spirit. Through their communication throughout the process, they helped me clearly understand their vision. They managed to make me share their excitement, which I believe was both effective and important.

Traction and Growth

As of 2025, Upsonic has reached an ARR level of USD 330,000. The platform is rapidly growing in the open-source community, with more than 7K stars, 200K+ downloads, and 1.2 million views on GitHub.

Closing Thought

In every period of transformation, some teams only follow the trend, while others set the standard. Our vision is to grow together with the teams that shape this standard. We believe that Upsonic will be one of the pioneers of this group. That is why we invested. P.S: We would like to thank Meryem Doğan for leading the investment process, Merve Kütükçüoğlu Karpuzcu for providing legal advice, and Burcu Oduncu for conducting the due diligence process.