Partner at Khosla Ventures
Check size: No personal check range is published. Khosla invests earliest from pre-seed/seed through Series A with conviction and also uses a Main Fund for later financings. Historical firm commentary described many seed investments around $1M–$2M; this is not Kanu's personal allocation range.
Kanu backs enterprise applications and infrastructure made possible by advances in AI, together with robotics and autonomous systems. Her technical and operating history spans hardware acceleration, predictive analytics, circuit simulation, and research, while her named portfolio covers conversational AI, world models, enterprise automation, navigation, weather, radar, warehouse robotics, autonomous trucking, and general-purpose robotics.
Go beneath the model demo. Explain the architecture, data advantage, performance envelope, failure modes, latency and cost, integration path, and how the system improves in deployment. For robotics, bring reliability, safety, hardware constraints, simulation-to-reality evidence, unit economics, and rollout milestones. Tie the technical breakthrough to a large enterprise or industrial budget and a durable feedback loop.
Research-grade technical depth paired with a practical product wedge; founders who can explain model or system advantage, data, hardware/software co-design, deployment constraints, and why their approach becomes economically superior at scale. Her portfolio suggests comfort with difficult, long-horizon engineering when milestones make risk legible.
The published focus implies caution toward undifferentiated AI applications, benchmark gains without customer value, robotics demos without reliability or deployment economics, infrastructure lacking a defensible advantage, and teams unable to specify technical risks, data loops, or the milestones required for commercial scale.
Connect a technical advance to a specific enterprise or industrial budget: quantify the workflow change, performance improvement, deployment burden, switching cost, and economic payback.
Evaluate perception, world model, planning, control, hardware, safety, reliability, simulation-to-real transfer, fleet learning, and service economics as one coupled system.
Ask whether real deployments generate proprietary data or operational feedback that measurably improves the model or system and widens the advantage faster than competitors can copy it.
+ 142 more investments. View fund →