Knuckle Up with Nakul · 2 min · Sep 9, 2026

Glean tests AI fluency and uses AI to screen hiring packets

Arvind Jain · CEO, Glean

Engineering candidates get a task designed to require AI, and an agent checks hiring packets against Arvind Jain's requirements.

In this episode

27:54 AI gives Jain strategic visibility he could not get before5 min
24:02 AI will raise customer expectations, not shrink company teamsPro pick 4 min
41:51 Fast innovation makes product moats feel temporaryPro pick 4 min
33:11 Leaders at Glean must build their own priority agents4 min
38:16Glean tests AI fluency and uses AI to screen hiring packets2 min
12:33 Core SaaS systems will remain, while point apps get rebuilt2 min
15:04 Glean says its context layer cuts token use by 30 to 50 percent4 min
3:54 Glean's edge depends on learning how enterprise work gets done3 min

“there's an agent that verifies every packet and they don't get routed to me until the agent actually clears it”

Arvind Jainat 39:45 · Knuckle Up with Nakul

Takeaways

Engineering tasks are too complex to finish in two hours without AI.
An agent checks hiring packets before they reach Jain.
AI fluency is a key check across departments.

About this segment

Engineering candidates get a task designed to require AI, and an agent checks hiring packets against Arvind Jain's requirements. The process changes how Glean assesses work and candidate feedback.

Arvind Jain is founder and CEO of Glean.

Founded Glean, the AI work assistant, after co-founding and leading R&D at Rubrik and a decade at Google.

Nakul Mandan, Host · Host, Knuckle Up with Nakul
General partner at Audacious Ventures; his show goes deep with founder CEOs on how they build.

From “$7B AI founder: "AI is not capable of replacing you right now" | Arvind Jain (CEO, Glean)”, Sep 9, 2026. Watch the full episode

Transcript

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

38:21
You talked about recruiting a little bit. Yeah, what has changed in your recruiting process beyond the thing you just mentioned? Is there a part of every interview has an AI and a development to it? Is there tasks you give? Certainly, like, you know, for example, for when we recruit engineers, so we used to have this programming exercise before. It was a two-hour session, and there was a program that we thought could be done if you write code yourself in two hours.
38:46
So we've changed that. Like now, we don't actually tell people and prompt them that you use AI to do it, but we give them a task which is so, so much more in complexity that there's no way for you to finish that in time in two hours, unless if you use AI, right? So that's sort of like, that's an example of how we change the interview process there. We've done that across departments. AI fluency is a key check.
39:09
But within the recruiting process, there are other things that we do. One of the things that, you know, I was very frustrated with was we use Greenhouse as our application technical system I still approve offers for employees, for every employee that joins the company and I would get these packets and I always felt that it was incomplete like it was missing like somebody puts in a feedback that oh I love this person, it's great, let's hire him and that's not a good enough feedback for me I would actually see what happened in the interview what questions were asked and so there was a lot of back and forth before and it was like my time would get spent and so now i actually said everything that i want to see in the packet is written in a document and now there's an agent that verifies every packet and they don't get routed to me until the agent
39:55
actually clears it you know within the system so this is an example of where AI is not replacing anyone, but it's just improving the quality of our work. Are you seeing, even within your organization, in every department, maybe it's slightly different,