Tarun Sachdeva

4.8K posts

Tarun Sachdeva banner
Tarun Sachdeva

Tarun Sachdeva

@tarunsachdeva

building https://t.co/T9YZWu3n99

Toronto Katılım Şubat 2008
1.8K Takip Edilen3.2K Takipçiler
Sabitlenmiş Tweet
Tarun Sachdeva
Tarun Sachdeva@tarunsachdeva·
i've basically dropped issue tracking the best way to see what our team is working on is to look at the traces
English
17
8
279
37.3K
sana
sana@greeningingreen·
@tarunsachdeva is one of the kindest and most generous people I’ve met in toronto, willing to take a shot on anyone, and putting his heart and soul into his projects and events i met him through the PROGRAM series he was running at @newsystems_ and he’s widened my understanding of what i think is possible through his determination rooting for you all thank you for everything :)
internetVin@internetvin

Starting today I’ll be helping @tarunsachdeva with @tracesdotcom in Toronto. Thank you for everything so far.

English
1
0
4
86
Tarun Sachdeva retweetledi
internetVin
internetVin@internetvin·
TRACES DOT COM
English
0
2
7
301
Tarun Sachdeva
Tarun Sachdeva@tarunsachdeva·
@sojoodi @xai Still playing around, terminal ux is pretty good so far, seems opentui based. will share some traces soon
English
0
0
1
26
Tarun Sachdeva
Tarun Sachdeva@tarunsachdeva·
i've basically dropped issue tracking the best way to see what our team is working on is to look at the traces
English
17
8
279
37.3K
0xSero
0xSero@0xSero·
I just received a 100,000$ grant from the Human Rights Foundation. In total I received: - 100K USD through HRF - 25.8K USD through donations site - 25K Brev credits through Nvidia - 4x B200s for a month - 5K from lambda - 4x RTX PRO 6000 private donor Open source must win
0xSero@0xSero

x.com/i/article/2034…

English
228
183
3.7K
176.4K
Tarun Sachdeva
Tarun Sachdeva@tarunsachdeva·
beyond tokenmaxxing the new traces contribution graph is a new view of your team's sessions over time, and a small step in giving you better tools to understand and analyze your sessions now live for all teams on @tracesdotcom
English
4
3
26
2.6K
Tarun Sachdeva
Tarun Sachdeva@tarunsachdeva·
playing with zero-native.dev tonight, this thing is crazy got a native mac app with webview in... 680kb
English
26
21
565
102.9K
zehra ✨
zehra ✨@zehranaqvi·
Introducing Annotations. Starting today, you can add field notes to your research in Lore with customization options like bold, italics, position, size, and color.
English
60
12
142
11.1K
Tarun Sachdeva retweetledi
Ben Vinegar
Ben Vinegar@bentlegen·
In 3 weeks at @devtoolsTO: @tracesdotcom - share agent coding sessions @polarityco - sandboxes for agent testing @super_user_app - agents in your team chat Tidra - AI code maintenance at scale Docsalot - docs agents can read Code Pathfinder - trace vulns across your codebase
Ben Vinegar tweet media
English
4
7
22
2K
Sam Hogan 🇺🇸
Sam Hogan 🇺🇸@samhogan·
We’re introducing HALO 😇 Hierarchal Agent Loop Optimizer HALO is an RLM-based agent optimization technique capable of recursively self-improving agents by analyzing their execution traces and suggesting changes. This work is inspired by the Mismanaged Genius Hypothesis proposed by @a1zhang and @lateinteraction earlier this month. tldr; we improved performance on AppWorld (Sonnet 4.6) from 73.7 --> 89.5 (+15.8) by giving HALO-RLM access to harness trace data and asking it to identify issues. The feedback from HALO surfaced failures in the harness such as hallucinated tool calls, redundant arguments in tools, refusal loops, and semantic correctness issues. Each issue mapped cleanly to a direct prompt update. We then fed these finding into Cursor (Opus 4.6), and asked the coding agent to update the underlying harness. We repeated this trace -> HALO-RLM analysis -> code update loop until the score plateaued. Today we’re open-sourcing the core HALO-RLM framework, evals, and data for further review.
Sam Hogan 🇺🇸 tweet media
English
59
124
1.4K
127.1K