Qodo Nabs $70M to Secure Code in the AI Era

As AI coding tools generate billions of lines of code monthly, a critical new bottleneck has emerged: ensuring the intended functionality of software. Qodo, an innovative startup developing AI agents for code review, testing, and governance, is making a significant bet that robust verification will be the defining characteristic of the next era in software development.

The New York-headquartered startup recently announced a successful $70 million Series B funding round, led by Qumra Capital, which propels its total capital raised to an impressive $120 million. The round saw participation from a diverse group of investors, including Maor Ventures, Phoenix Venture Partners, S Ventures, Square Peg, Susa Ventures, TLV Partners, Vine Ventures, along with notable individual investors Peter Welender (OpenAI) and Clara Shih (Meta).

Qodo positions itself as a vital layer designed to enhance trust in AI-generated code, especially as enterprises accelerate their adoption of sophisticated tools like OpenClaw and Claude Code. A common discovery among these organizations is that increased code output speed does not inherently guarantee the reliability or security of the resulting software.

Distinct from most AI review tools that primarily focus on identifying what has changed, Qodo's approach centers on analyzing how code alterations impact entire systems. It meticulously considers organizational standards, historical context, and risk tolerance, thereby empowering companies to manage their AI-generated code with greater confidence and precision.

Itamar Friedman, who previously co-founded Visualead and subsequently led the machine vision business at Alibaba (following its acquisition of Visualead), established Qodo in 2022. He shared with TechCrunch that his inspiration to launch Qodo, just months before the groundbreaking release of ChatGPT, stemmed from two pivotal moments in his career: his tenure at Mellanox (later acquired by Nvidia) and his experience building Visualead.

At Mellanox, where he contributed to automating hardware verification using machine learning, Friedman realized that “generating systems and verifying systems require very different approaches (different tools, different thinking).” Later, at Alibaba’s Damo Academy, he witnessed AI's evolution toward systems capable of sophisticated reasoning over human language. By 2021–2022, just prior to GPT-3.5, it became unequivocally clear to him that AI would be responsible for generating a substantial portion of the world’s content—particularly code—reinforcing his conviction that code generation and verification would necessitate fundamentally distinct systems.

A recent survey highlights a significant gap between awareness and practice: while a striking 95% of developers do not fully trust AI-generated code, only 48% consistently review it before committing.

“Code generation companies are largely built around LLMs. But for code quality and governance, LLMs alone aren’t enough,” Friedman asserted. He elaborated, “Quality is subjective. It depends on organizational standards, past decisions, and tribal knowledge. An LLM can’t fully understand that context. It’s like taking a great engineer from one company and asking them to review code at another — they lack the internal context.”

Friedman explained that while prominent companies like OpenAI and Anthropic are instrumental in shaping the broader AI narrative, including in adjacent areas such as code review, their primary focus tends to be on developing features rather than delivering comprehensive, end-to-end solutions. The CEO also noted that despite the presence of other startups in this space, many remain in early stages and have yet to achieve widespread enterprise adoption.

Qodo is strategically leveraging its superior performance to distinguish itself in an increasingly crowded market. The startup recently achieved the top position on Martian’s Code Review Bench, scoring an impressive 64.3% — more than 10 points ahead of its nearest competitor and a substantial 25 points clear of Claude Code Review. This benchmark performance emphatically demonstrates Qodo's capability to identify complex logic bugs and cross-file issues effectively, without overwhelming developers with excessive or irrelevant alerts.

In the past month, Qodo has launched Qodo 2.0, an advanced multi-agent code review system that is now leading current benchmarks. Concurrently, it has introduced innovative tools capable of learning and adapting to each organization’s unique definition of code quality.

The company has already forged strong partnerships with major enterprises, including NVIDIA, Walmart, Red Hat, Intuit, and Texas Instruments, in addition to high-growth firms such as Monday.com and JFrog.

“Every year has had a defining moment — from Copilot to ChatGPT to full task automation,” Friedman reflected. He concluded, “Now we’re entering a new phase: moving from stateless AI to stateful systems — from intelligence to ‘artificial wisdom.’ That’s what Qodo is built for.”