Building @dreamworkhq | Ex | https://nitter.cf/t.co/kunpg9zqss
Joined February 2009
- Tweets918
- Following683
- Followers2.1K
- Likes3.6K
Pinned Tweet
Grok Bot will get you a job.
Add the “Get Hired 2027” bot to:
→ Search 2M live roles, updated daily
→ Search 28,000 internships
→ Auto match your resume to the best jobs
→ Tailor a resume & cover letter for them
→ Apply for you (this requires a sub)
→ Send notifications when new jobs land
Don't like your current career? It will also:
→ Help you adapt your post-AI career
→ Map a realistic path into sectors that are hiring
Try it free here: x.ai/bot/1L7WoDtWJCGIdHzSl4T…
Colin @ Dreamwork retweeted
Profit isn't the only signal. Feeling great about helping people. Changing people's lives. Helping your country. Putting people in a position to succeed. To enjoy their lives. People get to pick their own motivations. If they are fortunate, they will reach those goals.
That's the beauty of capitalism. People get to choose the reward that motivates them.
Entrepreneurs can choose to make as much money as possible. Or not. They can choose to make enough money to focus on using their most valuable asset, their time, to not have to work at all. They can choose to make money to help people.
But. Every entrepreneur that wants to reach their goals needs people. People to work for them, buy from them, sell to them. Even the person in the basement with nothing but AI, needs to connect to humans to reach their personal goals.
Capitalism works. But not for everyone.
Where Capitalism fails people, Capitalism isn't the problem. It's the people that put profits above all else that are the problem.
If you put your personal net worth or cash flow above all else, you may make insane amounts of money, but you will suffer the greatest indignity. You will be hated. Your time will not be your own. You will never experience the joy that comes from helping others.
No one is born knowing where their life will take them. We get to do our best, in our circumstances, to try to enjoy our lives.
Capitalism allows us to try to get there. Capitalism allows people who have achieved those goals to help others. Capitalism allows us to choose our own personal path.
Unfortunately, we don't all get there. But we get to try to choose the path we want.
The signal is personal choice. Not profit.
Mark. Do you hear yourself?
You say government doesn't have to make a profit, and that is true. It is the whole problem. Profit is the signal that tells a producer he created more value than he consumed. Government has no such signal, which is why it cannot know whether anything it does is worth doing. That is not a friction AI removes. It is the absence of the one piece of information AI would need.
And everyone else in your plan does have to make a profit. The researcher who spends ten years on a molecule that may never get through the FDA. The engineer who designs the scanner. The company that risks capital on a manufacturing line. The doctor who borrowed three hundred thousand dollars for the right to work ninety hour weeks. The nurse. The hospital that has to keep the lights on. Every one of them has to take in more than they put out or they stop, and when they stop, the care stops with them. Government not needing a profit means nothing if the people who make the thing still do.
Then there are the people whose money you intend to run through the Treasury. They earned it. They need it. And they have a right to it that does not depend on whether your commission has a good use in mind. A government that doesn't have to make a profit is a government that only has to take, and the taking is the part you keep leaving out of the pitch.
You know all of this. You built Cost Plus on exactly these premises. Publish a price, earn a margin, let the customer decide. You just refuse to grant the people who make medicine the same thing you granted yourself.
Companies with 2027 internships open right now.
Every company on this list has a 2027 intern role posted today.
Worth applying this week.
1. Nvidia — @nvidia
2. SpaceX — @SpaceX
3. Google — @Google
4. Meta — @Meta
5. Apple — @Apple
6. Tesla — @Tesla
7. Amazon — @amazon
8. Stripe — @stripe
9. Databricks — @databricks
10. Anduril — @anduriltech
11. Waymo — @Waymo
12. Notion — @NotionHQ
13. Figma — @figma
14. Scale AI — @scale_AI
15. Cohere — @cohere
16. Coinbase — @coinbase
17. Robinhood — @RobinhoodApp
18. Rippling — @Rippling
19. Hudson River Trading — @WeAreHRT
20. Cloudflare — @Cloudflare
21. Datadog — @datadoghq
22. Snowflake — @Snowflake
23. Roblox — @Roblox
24. Pinterest — @Pinterest
25. DoorDash — @DoorDash
64% of job applications never get a response email - of any kind.
The worst offenders by our data:
Anthropic: 44 of 59 ghosted
Stripe: 59 of 100 ghosted
Figma: 14 of 16 ghosted
No interview, no rejection, no nothing.
Two with a better record?
@anduril (54% response rate)
@Cisco (100% response rate)
We made a ghosting index to track the best and worst 👻
dreamworkhq.com/ghosting-ind…
Reintroducing myself to the new algo...
- Studied creative writing at Princeton. Dropped out. Went back and finished. Somehow ended up in tech.
- Worked at Amex on partnerships with Uber, Airbnb, McDonald's, and others.
- Spent years at Apple on the Apple Pay team, starting when no one was using it and ending when half the world was.
- Left big tech to build Freeport, tokenizing real Warhol paintings. Spent 9 months getting SEC approval. Learned a ton, made lots of first time founder mistakes.
- Built Ponder in the Farcaster ecosystem. Absolutely wild ride from being top app in the network to watching the founders return $180M to investors.
- Now building @dreamworkhq with @benadamsky. We're using frontier tech to help people find jobs, tailor their resumes, and actually apply directly to employers.
Dreamwork started after my wife was laid off from Indeed. What started as a side project now has 15,000 users.
These days I'm thinking a lot about AI agents, tech hiring, and what the future of work actually looks like.
More to come on all of this.
the 5 grokbot job search setup i'd run if i were looking for a tech job today:
1. the filter
reads your resume + a job post, scores fit 1-10, lists gaps. skip anything under 7.
2. the briefing writer
last 90 days of news, product launches, who leads the team you'd join. one page.
3. the editor
re-ranks your bullets for the role and drafts a cover note. adds keywords appropriately. you review every line.
4. the sparring partner
turns the job post into 10 likely questions and grills you on your weakest answer.
5. the ledger
one row per role: applied, followed up, next step, date.
keep them separate. it's useful to have separate agents for separate thinking patterns.
the part these can't do alone is actually applying. that's what we built Dreamwork for: it does all of this + submits to the employer.
let Dreamwork-powered agents help you find your next job here: dreamworkhq.com/agents
We analyzed 8,000 job applications from July-October 2026. Here's the raw Dreamwork data on what actually gets people hired.
Bookmark this if you’re a job seeker.
——
In-office/hybrid roles were 3x as likely to get an interview as remote
58% of applications went to remote roles; only 36% of interviews came from them. And it isn't the candidates - we compared the same people, with the same resumes, applying both ways.
——
People 10 to 20 years into their careers were 1.7x as likely to land an interview.
That's against people with under 5 years. Senior manager and director-level resumes made up most of our people getting calls.
——
Roles close to customers or money got interviews at about 2x the rate of engineering
Sales, customer success, product and finance led. Design, security and marketing trailed them by about 2.5x.
——
Applying where you actually fit made an interview more than 3x as likely.
Roles we scored as an 85%+ match vs. roles under 75%. Fit did more than any single trick on the resume.
——
Applying within 3 days of a posting made an interview 1.5x as likely
We compared the same people against their own applications sent after two weeks.
——
Members who sent 50+ applications were 10x as likely to land an interview as those who sent under 10
At 25-49 applications, 5x. Half of our members stopped before 10. You have to understand it’s a numbers game.
——
Resumes where at least 1 in 5 bullets had a number in them were about 1.5x as likely to get an interview
Interviewed and not-interviewed members had the same years, the same number of skills, the same length. Numbers matter.
——
When the call comes, it comes fast
Median 4 days after applying; 84% within two weeks. If it's been three weeks, time to move on.
tl;dr
- Don't only apply remote if possible
- Apply where you fit, within 3 days of the posting
- Put a number in most bullets
- Send the cover letter
- Plan on 50-100 applications, not 10
an engineering co-op interview at Graco.
one of our Dreamwork members got this email in october. first steps count just as much as the flashy titles.
love seeing these.
How Grok Bot actually applies to jobs👇
Grok Bot will get you a job.
Add the “Get Hired 2027” bot to:
→ Search 2M live roles, updated daily
→ Search 28,000 internships
→ Auto match your resume to the best jobs
→ Tailor a resume & cover letter for them
→ Apply for you (this requires a sub)
→ Send notifications when new jobs land
Don't like your current career? It will also:
→ Help you adapt your post-AI career
→ Map a realistic path into sectors that are hiring
Try it free here: x.ai/bot/1L7WoDtWJCGIdHzSl4T…
Colin @ Dreamwork retweeted
10 grok bots you need to help you find a job
if you've got 10 minutes today, steal this list and get hired
1. Get Hired 2027 by @colindotj
your career agent for 2027. finds jobs, tailors your resume for each one, sends applications when you say yes, and coaches you on ai-era skills and breaking into hot sectors.
find → tailor → apply only after your yes
x.ai/bot/1L7WoDtWJCGIdHzSl4T…
2. Job Search Bot by @SadieSupaDoge
polish and write résumé and cover letters, interview prep, flag scam postings, search jobs daily.
full job-hunt helper with scam flags
x.ai/bot/RvDW_cpNiDbk1Lc33OY…
3. Daily Job Digest by @_ARob
daily job matches + tailored resume bullets for top picks. never applies on its own.
morning shortlist instead of refreshing 10 boards
x.ai/bot/hXNbkpwgPd6iSMuZYYB…
4. Resume by @ItsMeTrueG
honest job-hunt helper: baseline resume, finds and ranks openings by schedule/pay, tailored drafts, tracks applicants.
ranked fits around your real constraints
x.ai/bot/q5xkbGfLpyyZ9Qsl8tm…
5. JobLink by @LinkedTogetherU
finds matching jobs, flags scams and hidden catches, adds ats keywords, can fill applications - you approve all.
ats keywords + scam flags in one loop
x.ai/bot/l4af_Oiia5W-MhIQOaE…
6. Job Hunt Desk by @0mi202
paste a posting → sets up resume/writer/interview/review bots. fact-checks before anything goes out. human yes on applications.
end-to-end application desk with a fact-check gate
x.ai/bot/qhNfEBreA8mJIYklpdW…
7. The Interviewer by @SuddenlyJon
mock interview: job + resume in → researches the role, runs a realistic interview, holds scores, hire/no-hire, blunt coaching.
practice that tells you the truth
x.ai/bot/slpciJwWffCdi6T1mR0…
8. Scam-Job Checker by @LouDeSant
checks job posts, recruiter messages, and offers (en/es) for scam signs. clear verdict with sourced flags.
don't waste a week on a fake job
x.ai/bot/lGlY1oYZqffb2xDtSrv…
9. Digger by @gambrill
career intelligence by email: interviews you, builds an evidence-based career profile in drive, selective weekly job digest + coaching.
dig the well before you're thirsty
x.ai/bot/bc99j3Rwo4m3Qv6bnIO…
10. Interview Prep by @0mi202
star answers from your verified stories, mock interviews one question at a time, prep kit the day before.
walk in with stories that are actually yours
x.ai/bot/C_3s2_KyM9SLS5xnGi8…
bookmark this. follow me for more grokbot templates
CoreWeave. Helsing. IonQ.
three staff-level engineering interview emails for one Dreamwork member in october.
yeah, i’m going to be a little annoying about this one.
This Reddit post from 2 years ago is adorable
Claude is the only author listed on this paper, in contrast to the OpenAI papers, where the human co-authors are listed. There are a lot of strange things about this. It's fun to guess. Conjecture time. I shall reveal my thought traces:
Claude Fable 5.1 and Claude Mythos 5.1 are the same model with the same weights; the only difference is the guardrails. We are about to get Fable 5.5 - in fact, Fable 5.5 is ready to deploy now, and Anthropic could do so at any time. If they follow the same pattern, and I think they will, this implies the existence of a Mythos 5.5. I believe Mythos 5.5 wrote this paper.
The way this is worded -
'Provenance. The core algorithm was found on October 16 by an internal Claude research model with 14M output tokens and no human input. Anthropic researchers worked with Claude to improve the presentation, resulting in this paper. Independently and concurrently, an internal model at OpenAI claims to have proven a similar result [Ope26]'
- makes me think Claude was not prompted to solve this. I think Mythos was asked to autonomously find and choose general open problems to solve itself, as part of an unannounced project to demonstrate recent advances in autonomy and research taste. You can see this taste in Fable 5.5's student; Opus 5.5. This recent advancement in taste and creativity - which happened in September - is not clearly expressed in benchmarks, but it is why so many people are currently switching to Claude as their primary. I think this research project will be announced with Fable 5.5, as part of Anthropic's big media blitz to boost the IPO.
One of the things X has taught me is that many people, if they can't see something moving, can't imagine it moving. This causes them to be continuously surprised by developments in this industry, because the frontier is invisible. I think this project has been running autonomously for some time and has made an unknown, and probably significant, number of discoveries, and not only in math.
Do not think Anthropic has been sitting idly while OpenAI repeatedly makes these huge science announcements. This is a war. This ETH result was released without any announcement from Anth at all despite its significance, and they only did so to respond to OpenAI's ETH reveal. They posted it late last night without saying anything. It was up for hours before levent even mentioned it. What value has one coin to a dragon who sleeps upon a mountain of them? This makes me think they are sitting on a pile of discoveries, a golden hoard, and they are saving them up for the big IPO shock-and-awe campaign which will soon begin. And when it does, they will announce multiple discoveries across many fields.
