Instacart's CTO says AI made the company stop worrying about tech debt
Our take
Instacart’s recent shift towards AI-driven code generation and site reliability engineering presents a genuinely provocative challenge to conventional software development practices. The question posed by CTO Anirban Kundu – what if most of the work engineers do today could be handled by machines? – isn't simply a technological curiosity; it's a potential paradigm shift. The company’s experience, where builders now primarily navigate AI systems rather than directly writing code in 97% of cases, highlights a future where engineering focuses on higher-level orchestration, evaluation, and exception handling. This mirrors trends we've seen in the broader AI agent landscape, as explored in articles like Graph Engineering for AI Agents: Beyond the Single-Agent Loop, demonstrating the increasing complexity of agent architectures and their reliance on interconnected systems. The move also aligns with the broader conversation around modern AI agents and their connection to the real world, articulated in MCP Explained: How Modern AI Agents Connect to the Real World, highlighting the need for robust tool access and integration.
The implications of Instacart’s approach extend far beyond simply reducing tech debt. Their "Atoms" project, breaking down monolithic systems into modular components, echoes the principles of microservices architectures, albeit driven by AI rather than purely human design. The shift from code review to "intent model" training is particularly insightful. It suggests a future where engineers are less concerned with the minutiae of code syntax and more focused on defining desired outcomes and guiding the AI's problem-solving process. The success of their agentic SRE system, "Blueberry," which proactively identifies and mitigates production issues with significant accuracy, underscores the potential for AI to augment human expertise and dramatically improve operational efficiency. The ability of Blueberry to detect an AWS disk issue that human intuition initially missed speaks volumes about the comprehensive monitoring and pattern recognition capabilities of these systems—a capability that transcends our typical "first brain-second brain" debugging processes.
However, Instacart’s experience also serves as a cautionary tale. The 3% of cases where AI falls short—legacy systems, compliance requirements, and latency-sensitive workflows—demonstrate the continued importance of human expertise and the limitations of even the most advanced AI. The emphasis on democratizing code access by embedding domain knowledge in specifications, rather than relying on single "owner" teams, is a smart move towards fostering broader collaboration and reducing bottlenecks. This echoes the lessons learned from organizations like the payroll and HR software team, who successfully transitioned from monolithic systems to over 120 domain microservices, as detailed in Article: The Hard-Stop Rule: From 3 HCM Monoliths to 120 Domain Microservices. The key takeaway isn't the elimination of human engineers, but a fundamental reshaping of their roles and responsibilities.
Ultimately, Instacart’s experiment is pushing the boundaries of what’s possible in software development. The success they’ve achieved in leveraging AI to automate code generation and improve system reliability provides a compelling roadmap for other organizations seeking to unlock new levels of productivity and innovation. The question now becomes: how can organizations effectively integrate these AI agents into their existing workflows, cultivate the necessary skills in their engineering teams, and ensure that human oversight remains a critical component of the process—especially as AI models become increasingly complex and opaque? The future of software development likely hinges on our ability to answer this question effectively.
Instacart is posing the provocative question: What if most of the work your engineers do today should, in fact, be done by machines?
At VB Transform 2026, CTO Anirban Kundu argued that dev teams continue to waste their time on draining, repetitive, high-volume work; this should be absorbed by AI agents so that humans can focus on problems that require judgment, intent, and exception handling.
In fact, in 97% of cases, Instacart’s builders don’t even read code anymore.
“In the past, the tactical level was the creation of the code,” Kundu said. “In the most tactical level going forward, it's going to be, ‘How do you navigate around the AI system to give you what you want?’”
AI generating code, performing "pretty serious evals"
That doesn’t mean humans never look at code; agents handle the bulk of code generation and boilerplate, particularly with newer projects where code is generated or regenerated on a weekly basis.
“The benefit of that is we don't care about tech debt anymore,” Kundu said. “Things that are not active just get dropped out and then it gets rebuilt, kind of like how we used to build assembly code or object code.”
So why not 100%? The remaining 3% is in legacy, compliance, and latency-sensitive systems and workflows, or driven by a “boatload of code” that is dead, not active, or half-active. These cases still need careful human attention.
Instacart is slowly “smoothing those parts out,” however, breaking systems down in an aptly-named project Atoms, then building them back up in a cleaner, more modular form. Kundu’s team started with the “monoliths” and is shifting to remote procedure call (RPC)-driven architectures.
But evaluation remains one of the overarching challenges. Code reviews aren’t as relevant when AI is generating code — as Kundu noted, “the lines of code are going to be correct, the syntax is going to meet your expectations” — so the goal is to move to an “intent model.” That is, training devs so they can ask different models the right questions from an intent perspective.
Evals are then performed independently: Roughly 7,000 automatic evaluations run each month, and the system answers 8,000-plus real-time developer queries with about 99.9% accuracy.
Identifying "hiccups" that human intuition might have missed
Dovetailing with this, Instacart has built an agentic site reliability engineering (SRE) system trained on years of the company’s own incidents and root-cause analyses rather than generic failure data. Instead of teaching a model how production outages work in the abstract, the team fed it the specific ways Instacart’s systems have broken over time, along with the ways humans diagnosed and fixed them.
As a result, the company has seen accuracy in detecting and mitigating production issues jump from roughly 60 to more than 90%.
Kundu pointed to one example with Instacart’s internal tool Blueberry. The AI SRE colleague watches 200-some-odd Slack channels, monitors signals, and looks for patterns across human conversations and alerts.
In one incident, a database shard backed by an EBS volume that had a “hiccup” for a period of time. The human team did not immediately suspect AWS disk issues and were “obviously scrambling” to figure out why this particular shard misbehaved.
But about 20 minutes in, Blueberry posted on Slack, pointing to a specific blip and tying it to a feature-flag-like system called "roulette" that had been inadequate. "It's supposed to be rolling out in this cadence, [but] it had been too much,” Kundu said.
Blueberry figured it out, and the team resolved the incident. “Would have a human been as quick? I think the problem is human intuition would hold us back a little bit,” Kundu said.
Humans tend to default to patterns we’ve seen before, then resort to debugging; Kundu called this the “first brain-second brain kind of thing.” But Instacart’s agentic SRE is actually “more comprehensive in its ability to look at everything and then be able to decide what does or doesn't matter.”
Redefining the engineer’s job
Looking ahead, the most tactical work for engineers will be navigating AI systems: Designing and supervising evaluation processes; coordinating multiple simultaneous experiments and features; managing constraints like limited top-of-funnel traffic for testing; figuring out when to escalate; identifying edge cases and where things might break.
Domain expertise is also being rethought in the age of AI. Instead of bottlenecking changes through a single “owner” team that touches the code, Instacart is embedding domain knowledge into definitions and specs that any team can use.
“We’ve lived in this world where this group or this engineering team is the one that can touch the code and make the modification,” said Kundu. “We're trying to move into a world where the code becomes completely democratized across groups.”
Read on the original site
Open the publisher's page for the full experience