Samarth Aggarwal

Search & Intelligence @Glean. Formerly @meta AI, @adobe research, @quadeye, @IITDelhi CS, @UofIGrainger CS.

[Samarth Aggarwal @Samarth_A]

Glean is ~5x cheaper and ~3x higher quality than Claude Cowork. Enterprises no longer need to choose between what AI costs and what it delivers.
https://t.co/8kLp36P4sa

[Samarth Aggarwal @Samarth_A]

@sameenkarim @sameenkarim it would be amazing to have this. Is it already on the roadmap?

[Samarth Aggarwal @Samarth_A]

@sameenkarim feature req - pls add automerge for gh stacks

[Samarth Aggarwal @Samarth_A]

@svpino any tactical tips on what's worked for you to better design verifications for the overall system? ai-gen unit/integration tests are only one piece of the puzzle.

[Samarth Aggarwal @Samarth_A]

@neildhruva @d_sathe Ah, the missing piece is not telling it that the world will end if it broke out.

[Samarth Aggarwal @Samarth_A]

@sameenkarim when a pr is merged, I want the rest of the gh stack on top to automatically rebase on latest master w/o me having to manually run the sync. Is that possible today?

[Samarth Aggarwal @Samarth_A]

@DavidOndrej1 whats your setup like? how are you forcing subagent models and forcing handoffs for parallel execution and orchestration?

[Samarth Aggarwal @Samarth_A]

@pk_iv curious if you could expand on leveling up every other function pls? is it the clarity you get by refining the value with every pitch that helps you develop user empathy which is valuable in every function?

[Samarth Aggarwal @Samarth_A]

@thsottiaux Pls add per session cost tracking

[Aravind Srinivas @AravSrinivas]

At its peak, Sun Microsystems was valued at 205B (394B if inflation adjusted). Sold software in enterprise servers. Got disrupted by Linux, x86, and commodity hardware. Ended up selling to Oracle for 7.4B, losing 96% of its value.

Open source models running on local hardware can have a similar impact given what’s going on.

[Samarth Aggarwal @Samarth_A]

@athyuttamre @athyuttamre what are the key differences between live-1 and realtime-2? Curious why it is not realtime-3?

[Samarth Aggarwal @Samarth_A]

@athyuttamre @athyuttamre why can’t we select realtime-2 in the app?

[Samarth Aggarwal @Samarth_A]

@leodev Google secrets manager

[Samarth Aggarwal @Samarth_A]

@delba_oliveira Curious how you think about token efficiency here? Eg if the html generations cost as many tokens as the code while also not being the source of truth, isn’t it better to read the code diff directly?

[Samarth Aggarwal @Samarth_A]

@nicochristie one could say beyond a certain threshold of difficulty, latency stops mattering coz you're like take a full hour but solve this problem? arguably, mythos-level intelligence is only needed for tasks at that level of difficulty. wdyt?

[Samarth Aggarwal @Samarth_A]

@handotdev @mintlify insane flex

[Samarth Aggarwal @Samarth_A]

@rauchg so true, you need to be a software engineer to ensure AI agents can vibe code sustainably. What's a good metric to measure judgement for engineers and projects?

[Samarth Aggarwal @Samarth_A]

@snowmaker but this assumes users know and can express what they want, which is arguably harder than navigating a "well interfaced" website.

[Samarth Aggarwal @Samarth_A]

@beyang absolutely, even highlighting diff within a line

[Samarth Aggarwal @Samarth_A]

@thsottiaux managing git worktrees needs to be easier via codex