Kashish zooms out to discuss a universal industry pattern: how infrastructure—specifically data loading—is almost always the hidden constraint for ML scaling. The conversation dives deep into a recent architectural war story. Kashish walks through the full-stack profiling and detective work required to solve a massive GPU starvation bottleneck. By redesigning the Petastorm caching layer to bypass CPU transformation walls and uncovering hidden distributed race conditions, his team boosted GPU utilization to 60%+ and cut training time by 80%. Kashish also shares his philosophy on the fundamental trade-offs between latency and efficiency in GPU serving.

AI agents are shifting the role of developers from writing code to defining intent. This conversation explores why specs are becoming more important than implementation, what breaks in real-world systems, and how engineering teams need to rethink workflows in an agent-driven world.

The blog shows how to unify scattered multimodal assets, e.g., speaker bios, talk titles, videos, and PDFs. into a single, well‑structured memory layer.  It explains the metadata and schema decisions that let an agent answer richer, cross‑asset questions such as trending topics, influential speakers, or patterns across a full conference.  By grounding these relationships in a multimodal database like ApertureDB, the approach generalizes to any domain where organizations need AI to reason over diverse, real‑world collections of content.