There’s a rhythm to every revolution, and Deductive AI just dropped a track that hits all the right frequencies. Founded in June 2023 by Rakesh Kothari and Sameer Agarwal, this Mountain View startup is taking aim at the hidden tax of modern engineering, the hours lost debugging chaos across sprawling distributed systems. They just secured a $7.5M seed round led by CRV, with Databricks Ventures, Thomvest Ventures, and PrimeSet stepping in to back the vision.
Rakesh Kothari, once a founding engineer at Databricks and later at ThoughtSpot, knows large-scale data infrastructure like a jazz musician knows timing. Sameer Agarwal, Ph.D. from UC Berkeley’s AMPLab, helped architect Apache Spark and led Databricks’ Project Tungsten before managing 50+ engineers at Facebook as Area Tech Lead for Large-Scale data analytics. Now as CTO, he’s channeling that experience into something bolder, AI that thinks like an engineer and diagnoses systems like a doctor.
Deductive AI isn’t another observability dashboard; it’s a cognitive platform that connects code, logs, metrics, traces, and change events into a living knowledge graph. It doesn’t just flag an outage; it reasons through it. The platform’s AI agents learn from real-world incidents, generate hypotheses, test them, and surface root causes in seconds. Think of it as SRE meets Sherlock Holmes, powered by reinforcement learning and built on the Databricks Mosaic Research platform.
Early results are already serious. DoorDash used Deductive AI to automate debugging and cut incident resolution time by 90%, saving 1,000+ engineering hours and slashing root cause analysis by 80%. That’s time engineers now spend building, not firefighting, a quiet revolution in productivity that hits the bottom line.
The advisory lineup reads like a who’s who of tech’s elite: Ion Stoica of UC Berkeley and Apache Spark fame, Amit Singhal (ex-Google SVP of Search), Ajeet Singh (ThoughtSpot & Nutanix co-founder), Clint Sharp (CEO, Cribl), Ian Nowland (CEO, Junction Labs), and Abhinav Asthana (CEO, Postman). It’s less a board and more a brain trust built to scale brilliance.
This $7.5M round will accelerate product development, expand the engineering team, and deepen integrations with Prometheus, Datadog, New Relic, and OpenTelemetry. Backed by Databricks Ventures, Deductive AI also gains access to a network of 20K+ enterprise customers, including 60% of the Fortune 500.
Rakesh Kothari and Sameer Agarwal aren’t just solving outages; they’re building the intelligence layer for the modern cloud. In a world drowning in data and noise, Deductive AI doesn’t just find the signal; it understands the story behind it.
