Florian Juengermann
@florian_jue
Followers
2,347
Following
374
Media
32
Statuses
271
co-founder @listenlabs . prev cse @harvard , autopilot @tesla
Joined December 2018
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
Happy Pride
• 1692825 Tweets
Champions
• 1244991 Tweets
Real Madrid
• 1197192 Tweets
Dortmund
• 740109 Tweets
Vini
• 427504 Tweets
Kroos
• 406270 Tweets
Bellingham
• 235933 Tweets
Messi
• 235538 Tweets
مدريد
• 209340 Tweets
WNBA
• 177540 Tweets
Carvajal
• 175229 Tweets
Clark
• 152170 Tweets
Wilder
• 91239 Tweets
安田記念
• 81602 Tweets
Alex Jones
• 70188 Tweets
#UFC302
• 67950 Tweets
Dubois
• 48352 Tweets
الريال
• 44373 Tweets
Corinthians
• 42075 Tweets
#仮面ライダーガッチャード
• 40926 Tweets
#25thTGMA
• 34925 Tweets
Stonebwoy
• 28527 Tweets
#5vs5
• 28265 Tweets
Djokovic
• 25627 Tweets
#ブンブンジャー
• 21958 Tweets
ジンくん
• 21016 Tweets
Musetti
• 18524 Tweets
Botafogo
• 17153 Tweets
シャーシロ
• 13061 Tweets
Wesley
• 12283 Tweets
ニジゴン
• 11060 Tweets
We gave ChatGPT access to everything you know.
Every text that has been on your screen.
Every word spoken in a meeting.
(Powered by
@RewindAI
)
Now you can ask questions like:
- How do I know this person?
- What should be our OKRs this quarter?
Work with
@itsalfredw
30
151
871
While building our AI survey tool,
@itsalfredw
and I developed the fastest GPT4 audio pipeline on the planet*🤯
🗣️ Talk to the survey and it talks back!
💯 Try it out for yourself!
📝 Technical breakdown below.
43
79
416
We built an AI form 📋 that asks follow-up questions and makes form-filling fun.
No more feedback like “it’s broken” without any context 💁🏻♂️
🔍 Our AI digs deeper and learns why, so you don’t need to follow up.
Try Merlin 🧙🏼♂️
15
12
90
Breakdown of the latencies:
- 500ms transcription of the last audio chunk
- 500ms OpenAI latency
- 300ms until the first half-sentence is streamed
- 500ms text-to-speech latency
→ 1.8s total pause
Compare that to Pi...
7
0
45
Merlin is an AI-led user interview that asks follow-up questions and analyzes responses
👥 Instead of speaking to a few users, you can talk to everyone – instantly!
📞 Interview over phone, text or web
📊 Extract themes, sentiments, and personas
1
0
11
3
0
35
*at least I have not seen anything faster, let me know what you are using
Try it out here:
Voice playback currently only works on desktop.
4
4
31
You want a plane with a bed? I want a plane with a standing desk. We are not the same.
1
1
30
We are hiring for
@itsMerlinAI
. Do you know anyone that might be a good fit?
Merlin is hiring founding engineers! We give $2K to any recommendation that leads to a hire.
Why recommend/join Merlin?
- Our user numbers are growing like crazy!
- The best investors in Silicon Valley back us.
- We pay top-of-market SF salaries, with high equity + cash, and
90
27
523
3
0
29
Very soon LLMs will be able to answer any reasonable request.
What is missing?
Your knowledge. Your context.
With that you can answer questions like:
- How do I know this person?
- What should be our OKRs this quarter?
Limitless uses
@RewindAI
to get everything you’ve seen/heard and uses it to autocomplete any question.
They won 2nd place overall!
Built by
@florian_jue
,
@itsalfredw
(apologies - vid only last min of demo)
1
5
80
3
2
24
OpenAI officially turned on the money printer
Introducing ChatGPT Enterprise: enterprise-grade security, unlimited high-speed GPT-4 access, extended context windows, and much more. We’ll be onboarding as many enterprises as possible over the next few weeks. Learn more:
383
1K
5K
0
1
22
We have not launched yet, but sign up for early access here:
3
0
23
Having so much fun at our AI reading group at
@houseAtmosphere
We meet on a Tuesday night and discuss 2-3 influential AI papers.
In the past we had
@jeffreyzpan
@DevinShah16
@amanrsanger
@dluozhang
@danielmend_
present
4
1
22
hottest new innovation in AI: while (true)
now I spend $100/h on my GPT4 assistant that continuously explores my data and self-improves.
4
2
18
How we did it:
1. voice recording is streamed and transcribed in real-time
2. as soon as you stop, we send the transcript to GPT4
3. GPT4 streams the response back, token by token
2
0
16
At
@itsMerlinAI
we model our org structure after the transformer architecture
Just got a masterclass in leadership from Nvidia CEO Jensen Huang. Mind blown. Takeaways:
He has 55 direct reports, no 1:1s, just group talks = max transparency.
Org chart is modeled like a neural network, breaking the old pyramid model.
Nvidia bets on new grads to lead huge
214
608
7K
1
0
16
I had access to gpt-4-32k-0613 for a couple hours. Now it seems to be revoked again.
3
1
15
4. once there is a somewhat complete sentence, we send the first text chunk to a text-to-speech server
5. server streams synthetic voice back and we play it back as the bytes arrive
2
0
14
We won
@bentossell
's ChatGPT price
Winners 🏆
-
@marckohlbrugge
for marcbot, using Whisper API specifically. I hate telegram but just voice-noting with the bot was an A+ experience
-
@florian_jue
,
@itsalfredw
won the ChatGPT API challenge, they 'hacked' rewindai to build this mega handy assistant (check the
13
9
103
2
0
14
@AviSchiffmann
@itsalfredw
1. use streaming and start playing the sound as soon as the first bytes arrive
2. Elevenlabs has a "optimize_streaming_latency" parameter
3. (Most important) do not wait for the entire text but start sending the first words as soon as they come from GPT
2
1
13
Very cool demo!
Boundless + Limitless colab? 🤔
I’m creating a search engine that remembers everything on your screen and lets you ask abstract questions over all your data.
Not only that, I’ve developed AI that re-opens any indexed content in its original app.
Watch Boundless retrieve an Apple note with its virtual cursor:
52
89
1K
0
0
14
Bard just seems to make stuff up. Pretty disappointing experience so far...
8
0
14
Did anyone investigate if Falcon 40B has some biases eg regarding political views of the UAE?
If Falcon becomes the new open source standard, this will be crucial to know.
4
0
13
The robustness of Azure GPT4 has been critical to provide a reliable service to our customers. Glad to have this partnership with Microsoft!
AI is the ultimate disruptor. At
@itsMerlinAI
, they use Azure OpenAI Service to create GPT-4 powered surveys that engage like a human interviewer, while ingesting data securely on Azure.
Head to the blog to learn why Merlin chose Azure as their AI partner
0
4
16
0
1
13
A bit disappointed to see that the Gemini demo was not an actual video but some handpicked few-shot example with 3 images.
Now you just need to learn which 2 cups the hands touch, not really any temporal relation between the frames.
1
1
12
Have been playing around with fine-tuning a llama variant on a large document to get better hypothetical answers for retrieval.
Does not seem to beat one-shot GPT yet but it's a work in progress.
Let me know if you have thoughts on this.
1
1
12
I'm sure GPT *could* perform actions on websites like filling out forms, sending emails, etc.
The reason for the API approach for plugins is probably to have tighter control over which actions it is allowed to do.
1
1
11
Competitive programming is still far from being solved
2
0
12
@softyoda
@natfriedman
Based on the pricing on lambda ($2.40/h for a H100) it would run you roughly $144k on lambda.
0
1
12
Power outage in our office in SF since yesterday evening. Will be out till Saturday.
Almost 3 days without power in the financial district in one of the most advanced cities in the world?!
2
0
11
Limitless uses
@RewindAI
to get everything you’ve seen/heard and uses it to autocomplete any question.
They won 2nd place overall!
Built by
@florian_jue
,
@itsalfredw
(apologies - vid only last min of demo)
1
5
80
1
1
10
"Gemini was created from the ground up to be multimodal, highly efficient at tool and API integrations and built to enable future innovations, like memory and planning."
Will Google make agents work?
#GoogleIO2023
1
0
9
Super excited to try out
@maxvwolff
's new AI video editor!
Ozone raised $7.1m in seed funding and is launching its beta today!
Ozone is building the first AI-powered and collaborative video editing platform, designed to supercharge content creation for video creators and their teams.
Try Ozone now at .
38
73
839
1
0
9
@pranavrajpurkar
Me and my co-founder
@itsalfredw
(both from Harvard) are building
@itsMerlinAI
: collecting and analyzing qualitative feedback at scale.
Building our founding engineering team right now:
Merlin is hiring founding engineers! We give $2K to any recommendation that leads to a hire.
Why recommend/join Merlin?
- Our user numbers are growing like crazy!
- The best investors in Silicon Valley back us.
- We pay top-of-market SF salaries, with high equity + cash, and
90
27
523
0
0
8
I never understood why people used VPNs, but when I was back in Europe I set it up so I wouldn't have to see the GDPR cookie banners on every website..
0
0
8
What happens if you allow ChatGPT to execute arbitrary code? 🚨
@_alfredw
and me tried it out last weekend and built QueryGod.
1
0
7
Google is stopping all of its coding competitions including CodeJam and HashCode.
Is this to focus on their core business as competition is heating up?
20 years, more than a million participants and billions of lines of code later, our coding competitions are coming to a close. It's been an honor to learn and enjoy coding with you. Thank you.
Join us 4/15 at 2 p.m. UTC for farewell rounds of competition:
53
155
443
2
0
8
Hiring for
@itsMerlinAI
is going well..
Wow... Does anyone know the best way to filter over 5,000 job candidates? 😅
44
1
37
0
0
6
Bard thinks it's better than GPT-4 at generating creative and engaging text but worse at factual and informative text.
What are your experiences so far?
1
0
7
2-line autocomplete for any text box. LLMs will be used everywhere.
0
0
5
The future is now
⚡Introducing Bland Turbo. The world's fastest conversational AI:
- Send or receive up to 500,000+ phone calls simultaneously.
- Responds at human level speed. In anyone's voice.
- Program it to do anything.
Call and talk to Bland Turbo now:
🧵 (1/3)
241
329
2K
1
1
6
I always used to make fun and say "I'm pretty sure my code is correct, it's probably a bug in the compiler", today it was actually true.
1
0
5
Has anyone tried out streaming with Azure GPT-4?
Seems like the chunk size is much larger. Anyone knows a way around this?
I want it to have the smallest possible delay so it should send the first chunk immediately and not only after the first sentence.
1
0
5
Spent an evening trying to figure out why race conditions occur in my C++ code.
Turns out it's a bug in the gcc 11.1 stdlib.
2
0
5
@contentstrats
@RewindAI
@itsalfredw
Data storage and search are offline. However, we send the most relevant context info to ChatGPT once you trigger the shortcut to generate a response.
2
0
5
What has OpenAI discovered?
from sama YESTERDAY: "4 times now in the history of OpenAI - the most recent time was just in the last couple of weeks - I’ve gotten to be in the room when we push the veil of ignorance back and the frontier of discovery forward”
22
75
655
0
0
5
@karpathy
Makes sense. The SVM finds how your query is unique and then gives results that also have this unique property.
Will try it out on my data sets!
0
0
5
0
0
5
What does the next social media look like?
@_alfredw
,
@SilasAlberti
and me just launched our
@scale_AI
hackathon project:
🪄 ✨
Remix images with your friends
What’s the funniest thing you can create?
: Remix images with your friends ✨
@fjuengermann
,
@SilasAlberti
, and I built a new social network based on generative AI in 5 hours for the
@scale
hackathon.
Upload images and make funny edits with text.
9
17
124
0
0
4
Alphacode 2 powered by Gemini achieves a Codeforces ELO of ~1800, very impressive!
1
0
4
does anyone know the latencies and tps speeds for claude opus & sonnet on bedrock? Is it faster than antropics api?
2
0
4
Currently affected area. Good chunk of the financial district.
2
0
4
@yacineMTB
@altryne
@itsalfredw
Yeah, if you want to optimize for latency, GPT4 is probably not the way to go. But it has been pretty fast lately.
0
0
4
Interested? We would love to have you join the next session:
1
0
4
Our app BeFake was featured in an article in the Toronto star
At a glance, these pictures look just like me. But I never posed for them — they were entirely AI-generated, using cutting-edge technology from Google scientists called DreamBooth.
What could this mean for the future of deepfakes? My latest. (1/)
4
13
33
0
0
4
GPT-powered autocomplete for all websites!
How much time does it take to add AI to my friend's Webflow site? 🤖
2 minutes using EnhanceAI
Works now for any site! Check out the launch👇
4
6
41
0
0
3
This scene feels so weird...
Like a robot being thrown into the human world
0
0
4
Yesterday night other parts of the city were also affected. But seem to be restored after ~12h of outage.
0
0
3
1
0
3
0
0
3
@Eliasfiz
@itsalfredw
Use their streaming API. By the time you stop speaking, 95% is already transcribed!
3
0
3
@mharrvic
@itsalfredw
Hehe let’s say it’s still quite expensive right more. But costa will come down
0
0
3
Congrats
@mo_norouzi
on the launch!
We're excited to announce the formation of Ideogram AI today! Our mission is to help people become more creative through Generative AI.
69
148
642
1
0
3
@DavidSacks
Skipping 10 years ahead:
We gave ChatGPT access to everything you know.
Every text that has been on your screen.
Every word spoken in a meeting.
(Powered by
@RewindAI
)
Now you can ask questions like:
- How do I know this person?
- What should be our OKRs this quarter?
Work with
@itsalfredw
30
151
871
0
0
3
@quatermain32
@itsalfredw
You are right, reach was our main problem with the post. Need to have at least 5M impressions
1
0
3
0
0
3
the scary thing about the xz backdoor is that it was only discovered because they rushed the release and made mistakes (noticeable slow down and valgrind errors).
the valgrind errors were already fixed in the next release!
while this was already a 2 year effort, we will never
0
0
3
@joshm
@browsercompany
Better ordering of my recent tabs with command+T.
If I want to go back to the second most recently opened tab, command+T does not show it so I have to search through all of the tabs I have open on the left -- impossible!
I want to cycle through tabs by recency.
1
0
2
We gave ChatGPT access to everything you know.
Every text that has been on your screen.
Every word spoken in a meeting.
(Powered by
@RewindAI
)
Now you can ask questions like:
- How do I know this person?
- What should be our OKRs this quarter?
Work with
@itsalfredw
30
151
871
0
0
2
@fdotinc
: Remix images with your friends
: Remix images with your friends ✨
@fjuengermann
,
@SilasAlberti
, and I built a new social network based on generative AI in 5 hours for the
@scale
hackathon.
Upload images and make funny edits with text.
9
17
124
0
0
2
How can you make sure your LLMs work?
Let another LLM judge!
Unit test LLMs with LLMs
Tracking if your prompt or fine-tuned model is improving can be hard. During a hackathon,
@florian_jue
,
@fekstroem
, and I built “Have you been a good bot?”.
It allows you to ask another LLM to judge the output of your model based on requirements.
9
17
131
0
0
2
@DeveloperHarris
@itsalfredw
@Eliasfiz
it's pretty good! I had similar problems with it missing words at the end. Added some seconds of silence manually, that fixed it
1
0
2
1
0
2
@Jibran_05
@DeveloperHarris
@itsalfredw
@Eliasfiz
Yes , also implemented deepgram integration. But assembly is more accurate
0
0
2
@magicailabs
Do you have more I information on Long-Term Memory Networks? Like a write up, benchmark results etc?
0
0
2