Lightrun Secures $70 Million for AI-Powered Live Code Debugging

AI-powered coding has surged in popularity, promising to accelerate and simplify developers' work. However, this rise has also led to a massive increase in code volume, consequently raising the risk of bugs that cause crashes and other failures.

On Monday, Israeli startup Lightrun, which built an observability platform to identify and debug code before such problems occur, announced a $70 million Series B round. This funding highlights not only the market need for such tools but also Lightrun's success in addressing that demand.

New investor Accel is co-leading the round alongside existing backer Insight Partners, with participation from Citi, Glilot Capital, GTM Capital, and Sorenson Capital. To date, Lightrun has raised $110 million, including a Series A led by Insight that we covered in 2021.

The startup is not disclosing its valuation, but several indicators point to strong performance.

First, its client roster is impressive. Strategic backer Citi joins a list of major customers including ADP, AT&T, ICE/NYSE, Inditex, Microsoft, Priceline, Salesforce, and SAP.

Second, its product and market timing are advantageous. In July 2024, Lightrun launched a new AI-based debugging tool for use within integrated developer environments (IDEs), aptly named the Runtime Autonomous AI Debugger. While the company's platform was already effective, this product directly addresses a key enterprise challenge: AI is generating more code and, consequently, more problems. Lightrun built an AI tool to solve precisely that.

The company reports that revenue has grown more than fourfold since launch, attracting investor interest. Andrei Brasoveanu, the Accel partner who led the investment, stated he had been monitoring Lightrun for years and decided to invest following the new product launch.

Join the Disrupt 2026 Waitlist

Secure your spot on the Disrupt 2026 waitlist for priority access to Early Bird tickets. Previous Disrupt events have featured industry leaders from Google Cloud, Netflix, Microsoft, Box, Phia, a16z, ElevenLabs, Wayve, Hugging Face, Elad Gil, and Vinod Khosla — part of over 250 experts leading 200+ sessions designed to accelerate your growth. Additionally, connect with hundreds of startups pioneering innovation across all sectors.

Join the Disrupt 2026 Waitlist

Secure your spot on the Disrupt 2026 waitlist for priority access to Early Bird tickets. Previous Disrupt events have featured industry leaders from Google Cloud, Netflix, Microsoft, Box, Phia, a16z, ElevenLabs, Wayve, Hugging Face, Elad Gil, and Vinod Khosla — part of over 250 experts leading 200+ sessions designed to accelerate your growth. Additionally, connect with hundreds of startups pioneering innovation across all sectors.

"Everything aligned perfectly last year," he said. "Enterprise adoption accelerated significantly, all driven by AI."

CEO Ilan Peleg, who co-founded the company with CTO Leonid Blouvshtein, understands timing. Before pursuing further education and founding Lightrun, Peleg was a champion middle-distance runner, winning four national titles in Israel and ranking among Europe's top 16.

Peleg notes that dozens of companies currently build observability tools, with prominent players like Datadog and AppDynamics.

However, none have yet achieved the "holy grail": not only gaining a comprehensive view of all code deployed in production but also understanding how it interacts with existing systems, anticipating potential failure points, and doing so with minimal disruption and cost.

"Code is becoming cheap, but bugs are expensive," he remarked.

He added that this problem has reached "an inflection point." "Developers can now ship more code than ever before due to AI-driven automation. But fixing issues when they arise remains a highly manual process."

Lightrun's innovation is an observability toolset that monitors code within the IDE and predicts its behavior alongside live production code. The platform can then automatically adjust code as it deploys to production, preventing interruptions and crashes. It achieves this by running AI-based simulations to understand code behavior and preemptively fix issues.

"This is where we are unique," Peleg stated.

Given observability's close ties to other organizational functions, Lightrun has multiple potential growth paths.

One direction is developing tools tailored for cybersecurity teams, considering the security risks posed by bugs. Another is integrating its tooling closer to the point of code creation to make bug detection and resolution even more efficient.

For now, Peleg said the plan is to stay focused on enhancing its tools, talent, and business within the IDE. "We are mitigating everything that poses a risk to system resilience," he said, though he didn't rule out more specialized tools in the future.

Regarding code assistants, "they might be part of our future," he noted. "But even focusing solely on post-execution software remediation is a complex and vast challenge." Predicting the future of code creation is difficult, he added. Currently, with an estimated 30% to 60% of production issues stemming from code written by humans or machines, Lightrun is racing to provide a universal solution for observing and fixing all code, regardless of its origin.