# Harshul Jain Audible Talks LLM Inference at Scale
_Harshul Jain of Audible and Tanmay Sah were billed for an AI Engineer deep dive on LLM Inference at Scale, but source transcript yields no technical claims._
**Published:** 2026-09-08
**Source:** https://www.startuphub.ai/ai-news/artificial-intelligence/2026/harshul-jain-audible-talks-llm-inference-at-scale
---
[Audible](/startups/audible) principal Harshul Jain and independent AI researcher Tanmay Sah were billed for a deep dive on LLM Inference at Scale, according to [AI Engineer](https://www.youtube.com/watch?v=y2W4FNAuPEA) on September 8.

The Harshul Jain [Audible](/startups/audible) session title signals a focus on running large language models in production.

According to the source, the provided transcript contains only HLS playlist fragments and no extractable speech, slides, demos, or technical claims.

## How the Harshul Jain Audible session frames inference at scale

According to the source, no attack was demoed and no affected systems or attacker requirements were described.

There is no plain English exploit walkthrough to summarize and no analogy is warranted, because the source does not provide a vulnerability narrative, model, or tool chain.

## Why this matters and what is not fixed

For builders, inference at scale is about latency, throughput, cost, and reliability, not a disclosed security flaw here.

According to the source, no mitigations, patches, or builder guidance were stated, and no gaps were acknowledged on the record. Readers wanting the actual methods, numbers, or architecture choices will need the full video or slides.
---
Original analysis from [startuphub.ai](https://www.startuphub.ai), the #1 AI startup directory.