Lori Berenberg
@loriberenberg.bsky.social
investing early in the future of work at Bloomberg Beta. managed products at MongoDB and AppNexus.
This Bloomberg chart is the cleanest picture I’ve seen of circular financing in the AI era.
1/
1/
January 7, 2026 at 5:01 PM
This Bloomberg chart is the cleanest picture I’ve seen of circular financing in the AI era.
1/
1/
Usually, it’s pretty taboo to tell your boss you don’t like their work. Even stranger for them to ask for that feedback. Fortunately (or unfortunately?) for @roybahat.com, he asks, so I tell him exactly what I think… pretty much every day.
1/
1/
December 30, 2025 at 5:00 PM
Usually, it’s pretty taboo to tell your boss you don’t like their work. Even stranger for them to ask for that feedback. Fortunately (or unfortunately?) for @roybahat.com, he asks, so I tell him exactly what I think… pretty much every day.
1/
1/
I talked with 236 startups this year. And I noticed I ask a lot of the same questions.
On our Bloomberg Beta operating manual, we publish a list of our most common questions in pitch meetings.
I figured it might be interesting to share my own repeat offenders:
On our Bloomberg Beta operating manual, we publish a list of our most common questions in pitch meetings.
I figured it might be interesting to share my own repeat offenders:
December 22, 2025 at 5:04 PM
I talked with 236 startups this year. And I noticed I ask a lot of the same questions.
On our Bloomberg Beta operating manual, we publish a list of our most common questions in pitch meetings.
I figured it might be interesting to share my own repeat offenders:
On our Bloomberg Beta operating manual, we publish a list of our most common questions in pitch meetings.
I figured it might be interesting to share my own repeat offenders:
We tried an experiment at our quarterly team offsite two weeks ago: we turned on Granola, and we let our minds wander.
One of those conversations centered around a deceptively simple question… What makes a great founder?
1/
One of those conversations centered around a deceptively simple question… What makes a great founder?
1/
December 17, 2025 at 4:01 PM
We tried an experiment at our quarterly team offsite two weeks ago: we turned on Granola, and we let our minds wander.
One of those conversations centered around a deceptively simple question… What makes a great founder?
1/
One of those conversations centered around a deceptively simple question… What makes a great founder?
1/
The biggest difference between my job as a product manager vs. a startup investor? How I treat my calendar.
1/
1/
December 10, 2025 at 4:01 PM
The biggest difference between my job as a product manager vs. a startup investor? How I treat my calendar.
1/
1/
It's officially "the holidays."
Does that mean it's a terrible time to raise money for your startup?
You’ll get conflicting advice depending on who you ask, but the reality is, great founders will raise at any time of year.
1/
Does that mean it's a terrible time to raise money for your startup?
You’ll get conflicting advice depending on who you ask, but the reality is, great founders will raise at any time of year.
1/
November 29, 2025 at 12:00 AM
It's officially "the holidays."
Does that mean it's a terrible time to raise money for your startup?
You’ll get conflicting advice depending on who you ask, but the reality is, great founders will raise at any time of year.
1/
Does that mean it's a terrible time to raise money for your startup?
You’ll get conflicting advice depending on who you ask, but the reality is, great founders will raise at any time of year.
1/
When you’re demoing your product, put yourself in the shoes of the user. Even if it means showing something that isn’t your product.
1/
1/
November 12, 2025 at 5:01 PM
When you’re demoing your product, put yourself in the shoes of the user. Even if it means showing something that isn’t your product.
1/
1/
If you're a pre-seed startup, your pitch deck should resemble a local restaurant menu more than an NDA-filled McKinsey presentation.
I often hear founders worry about protecting their "most valuable" information until after they meet investors. That's backwards.
Here's why:
I often hear founders worry about protecting their "most valuable" information until after they meet investors. That's backwards.
Here's why:
October 22, 2025 at 4:04 PM
If you're a pre-seed startup, your pitch deck should resemble a local restaurant menu more than an NDA-filled McKinsey presentation.
I often hear founders worry about protecting their "most valuable" information until after they meet investors. That's backwards.
Here's why:
I often hear founders worry about protecting their "most valuable" information until after they meet investors. That's backwards.
Here's why:
lots of “drop your startup idea” threads act like VCs care primarily about the concept.
ugly truth about pitching VCs: the idea matters less than who you are and how you frame yourself as the right person to solve the problem.
ugly truth about pitching VCs: the idea matters less than who you are and how you frame yourself as the right person to solve the problem.
September 27, 2025 at 4:02 PM
lots of “drop your startup idea” threads act like VCs care primarily about the concept.
ugly truth about pitching VCs: the idea matters less than who you are and how you frame yourself as the right person to solve the problem.
ugly truth about pitching VCs: the idea matters less than who you are and how you frame yourself as the right person to solve the problem.
eleven labs: ai for audio
twelve labs: ai for video
...and they are not connected in any way?
twelve labs: ai for video
...and they are not connected in any way?
September 26, 2025 at 4:05 PM
eleven labs: ai for audio
twelve labs: ai for video
...and they are not connected in any way?
twelve labs: ai for video
...and they are not connected in any way?
the story goes like this...
startup a sells api tools to startup b
startup b sells hr software to startup c
startup c sells dev infra to startup a
all three backed by "tier 1" funds
all three report “growing enterprise adoption”
all three just raised at 100x arr
nothing to see here folks!
startup a sells api tools to startup b
startup b sells hr software to startup c
startup c sells dev infra to startup a
all three backed by "tier 1" funds
all three report “growing enterprise adoption”
all three just raised at 100x arr
nothing to see here folks!
September 25, 2025 at 5:00 PM
the story goes like this...
startup a sells api tools to startup b
startup b sells hr software to startup c
startup c sells dev infra to startup a
all three backed by "tier 1" funds
all three report “growing enterprise adoption”
all three just raised at 100x arr
nothing to see here folks!
startup a sells api tools to startup b
startup b sells hr software to startup c
startup c sells dev infra to startup a
all three backed by "tier 1" funds
all three report “growing enterprise adoption”
all three just raised at 100x arr
nothing to see here folks!
Reposted by Lori Berenberg
... with @angela-martin.bsky.social there since day one, and @mostlyfables.bsky.social an ally from the jump, and now working directly with us
... supported by team members past and present who we adore. @loriberenberg.bsky.social and others who have yet to fly in these blue skies.
... supported by team members past and present who we adore. @loriberenberg.bsky.social and others who have yet to fly in these blue skies.
June 5, 2025 at 7:27 PM
... with @angela-martin.bsky.social there since day one, and @mostlyfables.bsky.social an ally from the jump, and now working directly with us
... supported by team members past and present who we adore. @loriberenberg.bsky.social and others who have yet to fly in these blue skies.
... supported by team members past and present who we adore. @loriberenberg.bsky.social and others who have yet to fly in these blue skies.
Reposted by Lori Berenberg
When picking among the 9 AI models from OpenAI the rules are easy:
1) The model with the biggest number is mostly not the best
2) Mini means worse, except for the mini that is the second best
3) o1 pro beats o3-mini-high beats o1 beats o3-mini, naturally, except for creative work which is 4.5 or 4o
1) The model with the biggest number is mostly not the best
2) Mini means worse, except for the mini that is the second best
3) o1 pro beats o3-mini-high beats o1 beats o3-mini, naturally, except for creative work which is 4.5 or 4o
February 28, 2025 at 3:23 AM
When picking among the 9 AI models from OpenAI the rules are easy:
1) The model with the biggest number is mostly not the best
2) Mini means worse, except for the mini that is the second best
3) o1 pro beats o3-mini-high beats o1 beats o3-mini, naturally, except for creative work which is 4.5 or 4o
1) The model with the biggest number is mostly not the best
2) Mini means worse, except for the mini that is the second best
3) o1 pro beats o3-mini-high beats o1 beats o3-mini, naturally, except for creative work which is 4.5 or 4o
we are mere moments away from ChatGPT Stories
February 28, 2025 at 8:44 PM
we are mere moments away from ChatGPT Stories
with gpt-4.5, OpenAI has finally made a personality hire
February 28, 2025 at 8:44 PM
with gpt-4.5, OpenAI has finally made a personality hire
chatgpt has such a distinctively millennial voice
February 12, 2025 at 6:31 PM
chatgpt has such a distinctively millennial voice
at this rate, people will be building 401(k)s for AI agents before anyone in the real world actually uses AI agents
February 11, 2025 at 7:10 PM
at this rate, people will be building 401(k)s for AI agents before anyone in the real world actually uses AI agents
why does every email unsubscribe landing page look like it's from 1999
February 11, 2025 at 7:09 PM
why does every email unsubscribe landing page look like it's from 1999
this is the conversation i have every day with tech-forward people who have busy jobs. they’ve tried it all but no AI tooling has actually saved them enough time to be useful.
most people i know in the real world have zero patience for AI that's wrong/slow/unreliable!
most people i know in the real world have zero patience for AI that's wrong/slow/unreliable!
February 11, 2025 at 7:09 PM
this is the conversation i have every day with tech-forward people who have busy jobs. they’ve tried it all but no AI tooling has actually saved them enough time to be useful.
most people i know in the real world have zero patience for AI that's wrong/slow/unreliable!
most people i know in the real world have zero patience for AI that's wrong/slow/unreliable!
people are difficult. it's the reason vertical software exists, why enterprise sales cycles are long, why product managers have a job.
from a pure efficiency standpoint, all of these things could have been fixed or automated away already.
from a pure efficiency standpoint, all of these things could have been fixed or automated away already.
February 11, 2025 at 7:09 PM
people are difficult. it's the reason vertical software exists, why enterprise sales cycles are long, why product managers have a job.
from a pure efficiency standpoint, all of these things could have been fixed or automated away already.
from a pure efficiency standpoint, all of these things could have been fixed or automated away already.
an often unspoken truth: many successful people are fueled by spite
February 11, 2025 at 7:08 PM
an often unspoken truth: many successful people are fueled by spite
Reposted by Lori Berenberg
In a world of infinite slop, how do I figure out what to pay attention to?
February 4, 2025 at 5:04 PM
In a world of infinite slop, how do I figure out what to pay attention to?
27 really is the oldest age. i’m 29 and i’ve never felt younger
February 4, 2025 at 7:43 PM
27 really is the oldest age. i’m 29 and i’ve never felt younger
something i wish someone told me in high school or college: learning how to win over the “mean” teachers prepares you for the workplace far more than group projects, clubs, or leadership activities ever could
February 4, 2025 at 7:43 PM
something i wish someone told me in high school or college: learning how to win over the “mean” teachers prepares you for the workplace far more than group projects, clubs, or leadership activities ever could
devastating to listen to a recording of yourself, only to find out you've unknowingly adopted a new, even more annoying filler word
February 4, 2025 at 7:43 PM
devastating to listen to a recording of yourself, only to find out you've unknowingly adopted a new, even more annoying filler word