Manthan Gupta ✓

@manthanguptaa · SFO/ BLR · joined 01 Mar 2016

ai research engineer • designing agent runtimes, memory, evals & retrieval systems for autonomous agents • dms open

23 273Followers

360Following

14 322Posts total

285.8KViews on collected posts

Posts recentes

Manthan Gupta ✓@manthanguptaa12 Sep

Just saw someone use Google Search in a cafe in Bangalore. We are so early 33.8K views · 569 likes · 0 reposts · 7 replies Open on X →

Manthan Gupta ✓@manthanguptaa10 Sep

I found the correct use of iphone duo https://t.co/py3P5U3ix1

18.9K views · 207 likes · 4 reposts · 8 replies Open on X →

Manthan Gupta ✓@manthanguptaa10 Sep

Great article. It’s a really good complementary read to one of my posts on using LLMs as a judge: https://t.co/V376LhGRV5 One thing I would take away from the article is that for long-running agentic tasks, we probably shouldn’t judge only the final output. We should judge the 13.6K views · 156 likes · 8 reposts · 7 replies Open on X →

Manthan Gupta ✓@manthanguptaa9 Sep

The last one was so good that we had to do it again. We are back with the next edition of Research on Tap, and this time we are diving into inference engineering. 10 people around one table, discussing a few carefully picked papers and going deep on what actually makes modern h

1.7K views · 11 likes · 2 reposts · 2 replies Open on X →

Manthan Gupta ✓@manthanguptaa6 Sep

It was a privilege to spend 3.5hrs with these people to discuss and disect the HELIX research paper and also talk about recursive self improving agents. Research on Tap S2 with @udayan_w. Lookout for the next one soon What should be the topic of the next one? https://t.co/d8Jz3

0:10

29.9K views · 108 likes · 2 reposts · 4 replies Open on X →

Manthan Gupta ✓@manthanguptaa15 Aug

https://t.co/hf076ycRDY 20.8K views · 96 likes · 8 reposts · 13 replies Open on X →

Manthan Gupta ✓@manthanguptaa9 Apr

@manthanguptaa @steipete I’m just saying 4 views · 0 likes · 0 reposts · 0 replies Open on X →

Manthan Gupta ✓@manthanguptaa29 Jan

@nikitabier my entry for the $1M pot for best articles. Only if you allowed it in India 😮‍💨🫠 2.1K views · 7 likes · 0 reposts · 0 replies Open on X →

Manthan Gupta ✓@manthanguptaa27 Jan

Checkout more blogs from me https://t.co/o4bCE3Cr1w 58.3K views · 13 likes · 1 reposts · 0 replies Open on X →

Manthan Gupta ✓@manthanguptaa26 Jan

@manthanguptaa So essentially the same thing Claude-mem uses. Adding memory_search tool doesn’t work well for LLMs because they aren’t trained to search the memory before thinking, they never do this. You have to be proactive with memory and load everything necessary into context 20K views · 21 likes · 0 reposts · 1 replies Open on X →

Manthan Gupta ✓@manthanguptaa26 Jan

Thanks for building clawdbot @steipete 🙌 86.7K views · 59 likes · 0 reposts · 1 replies Open on X →

Em comparação com contas do mesmo porte

6 posts dos últimos 90 dias, ao lado da faixa de 10K–100K seguidores. aparece para muita gente, mas poucos desses espectadores reagem.

Mediana de visualizações19 862esta conta953mediana para 10K–100K

Alcance, %85.34%esta conta3.62%mediana para 10K–100K

Engajamento, %1.01%esta conta1.51%mediana para 10K–100K

Métrica	Esta conta	Mediana para 10K–100K	Proporção
Mediana de visualizações por post	19 862	953	20.8×
Alcance (visualizações ÷ seguidores)	85.34%	3.62%	23.6×
Taxa de engajamento	1.01%	1.51%	0.67×

Outras contas desta faixa → Comparar com outra conta → Como estas referências são construídas →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

86.7K26 Jan

20K

58.3K27 Jan

2.1K29 Jan

49 Apr

20.8K15 Aug

29.9K6 Sep

1.7K9 Sep

13.6K10 Sep

18.9K

33.8K12 Sep

Last 11 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

0.07%26 Jan

0.11%

0.02%27 Jan

0.33%29 Jan

0.00%9 Apr

0.56%15 Aug

0.38%6 Sep

0.86%9 Sep

1.26%10 Sep

1.16%

1.71%12 Sep

Reactions — likes, reposts, replies and quotes — divided by views. Median for 10K–100K accounts is 1.51%.

What the audience does

Likes72.3%1 247 in total

Reposts1.5%25 in total

Replies2.5%43 in total

Bookmarks23.7%409 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.