holy shit lmao
fable 5.2 made this for me in ~2 hours https://t.co/rK2p3oneBW
Jarrod Watts
1D
TODAY
7D
WTD
30D
MTD
90D
QTD
12M
YTD
Activity
Tokens tagged by the author per day over the selected time range. The most mentioned tokens have their own color, the rest add up in Other. A post that tags two tokens counts for both.
MENTIONS
ACTIVITY
TOTAL
Token Mentions
Tokens the author tagged in posts over the selected time range, most mentioned first: posts per token, the share of its classified posts that was bullish and the average engagement score of those posts, with price changes, market cap and volume as in the screener, plus price and sector on wide screens.
Latest Posts
The author's newest posts in the selected time range.
@jarrodwatts
I completely forgot Fable 5.1 was a model tbh
Have been exclusively using Opus 5.5 since it dropped
11/9/2026 23:16
2.6K
6
0
52
1
@jarrodwatts
I spent months of my life and thousands of dollars looking for the best AI coding setup.
I've finally found it - T3 Code.
So, I made a video showing my full $1,000/mo setup: https://t.co/km5qFt1W66
11/8/2026 03:45
54K
60
30
527
472
@jarrodwatts
I asked Astra to edit Qwen3.8-27B's weights and make it more token efficient.
For ~$100 in GPU cost, it found the weights for the "done thinking" signal and edited them to be twice as strong.
The result: 40% fewer thinking tokens on average on the ...
11/4/2026 02:39
109K
105
80
1.7K
826
@jarrodwatts
Opus 5.5 one-shotted a replica of this Linux box that I bought using Blender https://t.co/4eDneaMPm6
11/2/2026 23:43
2.4K
5
0
27
0
@jarrodwatts
how can anyone not love ai https://t.co/bXhm1Kvvn1
11/2/2026 02:44
3K
3
1
28
1
Lead AI Engineer @monad
| Handle: | @jarrodwatts |
| Location: | Sydney, Australia |
| Website: | www.jarrodwatts.com |
| Joined: | February 2021 |
Author Data
The author's audience and activity. Followers change over the selected time range. Posts and Avg Engagement (the average engagement score per post, as in the feed's Engagement Score sort) cover the selected time range. Engagement Rank ranks the authors who posted in the range by average engagement score. All of the author's posts count: the filter chips do not apply here.
Followers 34K +1.15% (30D)
Following 1.7K
Posts (30D) 18
Avg Engagement (30D) 13
Engagement Rank #76 of 1264
Sentiment
Share of the author's classified posts that were bullish, neutral and bearish over the selected time range. Posts without a sentiment classification are not counted. All of the author's posts count: the filter chips do not apply here.
Top Post
The author's post with the highest engagement score in the selected time range.
@jarrodwatts
I asked Astra to edit Qwen3.8-27B's weights and make it more token efficient.
For ~$100 in GPU cost, it found the weights for the "done thinking" signal and edited them to be twice as strong.
The result: 40% fewer thinking tokens on average on the ...
11/4/2026 02:39
109K
105
80
1.7K
826