Subj : Nvidia GPU cloud ¨, DeepMind CEO interview ¨, Carta CTO learnings ¨¨ To : All From : TLDR Date : Mon May 26 2025 10:44:03 --kTz8PYjU Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Nvidia's DGX Cloud Lepton is a new service designed to link artificial in= telligence developers with Nvidia's network of cloud providers=C2=A0= =E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2= =A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C= =C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80= =8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0= =E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2= =A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0= =E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2= =A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C= =C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80= =8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0= =E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2=A0=E2=80=8C=C2= =A0 Sign Up [1] |Advertise [2]|View Online [3]=20 =09=09TLDR = =09=09TOGETHER WITH [WorkOS] [4] TLDR 2025-05-26 WORKOS AU= THKIT + RADAR: AUTHENTICATION WITH REAL-TIME THREAT PROTECTION (SPONSOR) = [4]=20 Authentication isn't just login. It's building secure flows, man= aging sessions, handling tokens, supporting SSO, and protecting user data= .. AuthKit [5] provides a fully hosted, customizable UI for sign-up, pas= sword resets, MFA, and seamless SSO =E2=80=94 production-ready from day o= ne. But authentication alone doesn't stop abuse. WorkOS Radar [6] adds= real-time threat prevention, detecting brute force, credential stuffin= g, and more. Built-in protections can be enabled instantly =E2=80=94 no s= cripts or custom logic required. Build secure, scalable authentication= from day one with AuthKit and WorkSO Radar. [7] =F0=9F=93=B1=20 = BIG TECH & STARTUPS THE MAN WHO =E2=80=98A.G.I.-PILLED' GOOGLE (56 MI= NUTE READ) [8]=20 Demis Hassabis, the chief executive of Google DeepMin= d, has been dreaming of artificial general intelligence (AGI) for years. = He was already taking the possibility of AGI seriously when Google acquir= ed DeepMind in 2014. Today, he is among a growing number of tech leaders= racing to build AGI. Hassabis and a colleague received the Nobel Prize= in Chemistry last year for their work on AlphaFold, an AI system capab= le of predicting the three-dimensional structures of proteins. This artic= le contains a transcript of an interview with Hassabis about his views on= AGI and the future that follows its arrival.=20 NVIDIA PUSHES FURTHER = INTO CLOUD WITH GPU MARKETPLACE (2 MINUTE READ) [9]=20 Nvidia's DGX C= loud Lepton is a new service designed to link artificial intelligence dev= elopers with Nvidia's network of cloud providers, including CoreWeave, La= mbda, and Crusoe. It provides access to Nvidia's graphics processing unit= s, making those chips more widely accessible to all developers. The platf= orm will have a marketplace of GPU cloud vendors that developers can pick= from to train and use their AI models. Developers can choose which cloud= provider to work with and even work with multiple cloud providers.=20 = =F0=9F=9A=80=20 SCIENCE & FUTURISTIC TECHNOLOGY ROBOTS ARE STAR= TING TO MAKE DECISIONS IN THE OPERATING ROOM (11 MINUTE READ) [10]=20 = Autonomous robots aren't being used to operate on people yet, but the te= chnology is almost capable of such a feat and more autonomy is on the way= .. The technology could help hospitals meet the high demand for safety and= consistency during surgeries, manage routine tasks, and prevent mistakes= .. Future robots will use AI to autonomously manage movements and decision= -making during surgical tasks, reducing the need for constant human input= .. With the number of surgeons around the world declining while the number= of people who need surgery continues to increase, the need for the techn= ology is clear.=20 VALVE CEO GABE NEWELL'S NEURALINK COMPETITOR IS EXPE= CTING ITS FIRST BRAIN CHIP THIS YEAR (3 MINUTE READ) [11]=20 Starfish= Neuroscience, a brain-computer interface startup incorporated by Valve c= o-founder and CEO Gabe Newell, plans to produce its very first brain chip= later this year. The company is currently seeking collaborators to open = up avenues for the technology. Starfish's goal is to create a smaller and= less invasive implant than the competition that can enable simultaneous = access to multiple brain regions. Details from the chip's current spec sh= eet are available in the article. The article also links to a video of a = talk from Valve that explains how the technology could enhance gaming. = =F0=9F=92=BB=20 PROGRAMMING, DESIGN & DATA SCIENCE STUFF I L= EARNED AT CARTA (7 MINUTE READ) [12]=20 Will Larson worked at Carta as = its CTO for the past two years. This post contains the most important les= sons Larson learned in his time at the company, covering topics such as e= ngineering strategy, large language model adoption, multi-dimensional tra= deoffs, creating software quality, and more.=20 OPENAI: SCALING POSTG= RESQL TO THE NEXT LEVEL (18 MINUTE READ) [13]=20 Bohan Zhang, a member = of OpenAI's Infrastructure team, recently shared the company's best pract= ices with PostgreSQL at the PGConf.dev 2025 Global Developer Conference. = The talk offers a glimpse into the database usage of one of the most prom= inent unicorn companies and demonstrates how PostgreSQL can scale gracefu= lly under massive read loads. OpenAI utilizes managed databases on Azure,= employing a classic PostgreSQL primary-replication architecture without = sharding. This article covers the content of Zhang's talk along with comm= ents and includes some more details about OpenAI's infrastructure.=20 = =F0=9F=8E=81=20 MISCELLANEOUS AT AMAZON, SOME CODERS SAY THEIR JOB= S HAVE BEGUN TO RESEMBLE WAREHOUSE WORK (12 MINUTE READ) [14]=20 As A= I spreads through the labor force, many white-collar workers say their wo= rk is becoming more routine, less thoughtful, and much faster-paced. Comp= anies seem to be persuaded that AI can increase productivity. Teams are g= etting smaller, but they're being expected to output the same amount of w= ork by using AI. Managers say that AI can relieve employees of tedious ta= sks and enable them to perform more interesting work, but for inexperienc= ed workers, the change resembles the shift from artisanal work to factory= work in the 19th and 20th centuries.=20 ASK HN: GO DEEP INTO AI/LLMS= OR JUST USE THEM AS TOOLS? (HACKER NEWS THREAD) [15]=20 Large langua= ge models are becoming commodities. Many people use them, but few people = are developing them from scratch. Developing large language models involv= es a lot of hardcore computer science and math - people can use them with= out having to know any of the internals. While it might be helpful to und= erstand some of the internals and math, it's probably best to stick to le= arning how to get good at using these models and integrating what they do= into agentic systems.=20 =E2=9A=A1=20 QUICK LINKS LOVING TLDR = BUT WANT MORE AI? (SPONSOR) [16]=20 We also write TLDR AI. Same format,= all about LLMs, GenAI, and deep learning. Delivered every weekday. Subsc= ribe for free. [16]=20 ROOT FOR YOUR FRIENDS (4 MINUTE READ) [17]=20 = Rooting for your friends is the best way to improve your career, happ= iness, and the people you care about.=20 F2 (GITHUB REPO) [18]=20 F= 2 is a cross-platform command-line tool for batch renaming files and dire= ctories quickly and safely.=20 REINVENT THE WHEEL (3 MINUTE READ) [19] = You should learn from others, but you also need to put your knowledg= e to the test so you learn enough about your field to advance it.=20 = ON FILE FORMATS (3 MINUTE READ) [20]=20 This post presents things to co= nsider if you are thinking about creating a new file format.=20 TRAIL= BASE (GITHUB REPO) [21]=20 TrailBase is an application server with sub-= millisecond latencies, eliminating the need for dedicated caches.=20 = MCP IS THE COMING OF WEB 2.0 2.0 (5 MINUTE READ) [22]=20 MCP could insp= ire a new generation of developers to push for the web to go back towards= its natural architecture, programmable through janky specs that everybod= y adopted for the sheer joy of something fun to hack on.=20 Love TLDR?= Tell your friends and get rewards! Share your referral link below wi= th friends to get free TLDR swag!=20 https://refer.tldr.tech/1fd848fa/ = [23]=20 =09=09 Track your referrals here. [24]=20 Want to advertise i= n TLDR? =F0=9F=93=B0 If your company is interested in reaching an aud= ience of tech executives, decision-makers and engineers, you may want to = ADVERTISE WITH US [25].=20 Want to work at TLDR? =F0=9F=92=BC AP= PLY HERE [26] or send a friend's resume to jobs@tldr.tech and get $1k if = we hire them!=20 If you have any comments or feedback, just respond to = this email!=20 Thanks for reading,=20 Dan Ni [27] & Stephen Flanders [28= ]=20 Manage your subscriptions [29] to our other newsletters on tech,= startups, and programming. Or if TLDR isn't for you, please unsubscrib= e [30].=20 =20 Links: ------ [1] https://tldr.tech/signup?utm_sou= rce=3Dtldr [2] https://advertise.tldr.tech/?utm_source=3Dtldr&utm_medium= =3Dnewsletter&utm_campaign=3Dadvertisetopnav [3] https://a.tldrnewsletter= ..com/web-version?ep=3D1&lc=3Db5bb39ac-a04f-11ee-8692-6388dbdd6fd1&p=3D1c9e1= b52-3a06-11f0-90a0-1980b8e627ce&pt=3Dcampaign&t=3D1748256243&s=3De32865ba3e= e4af98b10a3dc869d72211b3dff8cfb008d256dc1519aad803ff74 [4] https://workos= ..com/blog/how-workos-radar-really-works?utm_source=3Dtldr&utm_medium=3Dnews= letter&utm_campaign=3Dq22025 [5] https://www.authkit.com?utm_source=3Dtld= r&utm_medium=3Dnewsletter&utm_campaign=3Dq22025 [6] https://workos.com/ra= dar?utm_source=3Dtldr&utm_medium=3Dnewsletter&utm_campaign=3Dq22025 [7] h= ttps://workos.com/docs/user-management/authkit?utm_source=3Dtldr&utm_medium= =3Dnewsletter&utm_campaign=3Dq22025 [8] https://links.tldrnewsletter.com/= fvM811 [9] https://links.tldrnewsletter.com/J26a20 [10] https://spectru= m.ieee.org/star-autonomous-surgical-robot?utm_source=3Dtldrnewsletter [11= ] https://www.theverge.com/news/673938/gabe-newell-valve-founder-brain-comp= uter-interface-first-chip-starfish?utm_source=3Dtldrnewsletter [12] https= ://lethain.com/stuff-learned-at-carta/?utm_source=3Dtldrnewsletter [13] h= ttps://www.pixelstech.net/article/1747708863-openai%3a-scaling-postgresql-t= o-the-next-level?utm_source=3Dtldrnewsletter [14] https://links.tldrnewsl= etter.com/Gi1o4O [15] https://news.ycombinator.com/item?id=3D44079303&utm= _source=3Dtldrnewsletter [16] https://tldr.tech/ai/?utm_source=3Dtldr&utm= _medium=3Dnewsletter&utm_campaign=3Dquicklinks05262025 [17] https://josep= hthacker.com/personal/2025/05/13/root-for-your-friends.html?utm_source=3Dtl= drnewsletter [18] https://github.com/ayoisaiah/f2?utm_source=3Dtldrnewsle= tter [19] https://endler.dev/2025/reinvent-the-wheel/?utm_source=3Dtldrne= wsletter [20] https://solhsa.com/oldernews2025.html#ON-FILE-FORMATS?utm_s= ource=3Dtldrnewsletter [21] https://github.com/trailbaseio/trailbase?utm_= source=3Dtldrnewsletter [22] https://www.anildash.com//2025/05/20/mcp-web= 20-20/?utm_source=3Dtldrnewsletter [23] https://refer.tldr.tech/1fd848fa/= [24] https://hub.sparklp.co/sub_32edf8dd66ae/1 [25] https://advertise.= tldr.tech/?utm_source=3Dtldr&utm_medium=3Dnewsletter&utm_campaign=3Dadverti= secta [26] https://jobs.ashbyhq.com/tldr.tech [27] https://twitter.com/= tldrdan [28] https://twitter.com/SteveFlanders22 [29] https://tldr.tech= /tech/manage?email=3Dtldr%40synchro.net [30] https://a.tldrnewsletter.com= /unsubscribe?ep=3D1&l=3Dcfa2d55a-b7be-11e8-a3c9-06b79b628af2&lc=3Db5bb39ac-= a04f-11ee-8692-6388dbdd6fd1&p=3D1c9e1b52-3a06-11f0-90a0-1980b8e627ce&pt=3Dc= ampaign&pv=3D4&spa=3D1748253652&t=3D1748256243&s=3D6b9fcf2d0df9fcd7ca2c24f4= aff5e27fde5f51c2af5bf63d5b4c427ae739ec01 --kTz8PYjU Content-Type: text/html; charset=utf-8 Content-Transfer-Encoding: quoted-printable TLDR
Nvidia= 9;s DGX Cloud Lepton is a new service designed to link artificial intellige= nce developers with Nvidia's network of cloud providers =E2=80=8C=  =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C=  =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C=  =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C=  =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C=  =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C=   =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2= =80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2= =80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2= =80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2= =80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2=80=8C =E2= =80=8C =E2=80=8C 

TLDR

<= /tr>
Together With

TLDR 2025-05-26

WorkOS AuthKit + Radar: Aut= hentication with Real-Time Threat Protection (Sponsor)

Authentication isn't just login. It's b= uilding secure flows, managing sessions, handling tokens, supporting SSO, a= nd protecting user data. AuthKit= provides a fully hosted, customizable UI for sign-up, password resets, MFA= , and seamless SSO =E2=80=94 production-ready from day one.

But authentication alone doesn't stop abuse. WorkOS Radar adds real-time threat prevention, detecting= brute force, credential stuffing, and more. Built-in protections can be en= abled instantly =E2=80=94 no scripts or custom logic required.

Build secu= re, scalable authentication from day one with AuthKit and WorkSO Radar.

=F0=9F= =93=B1

Big Tech & Startups

The Man Who =E2=80=98A.G.I.= -Pilled' Google (56 minute read)

Demis Hassabis, the chief executive of = Google DeepMind, has been dreaming of artificial general intelligence (AGI)= for years. He was already taking the possibility of AGI seriously when Goo= gle acquired DeepMind in 2014. Today, he is among a growing number of tech = leaders racing to build AGI. Hassabis and a colleague received the Nobel Pr= ize in Chemistry last year for their work on AlphaFold, an AI system capabl= e of predicting the three-dimensional structures of proteins. This article = contains a transcript of an interview with Hassabis about his views on AGI = and the future that follows its arrival.
Nvidia Pushes Further Into = Cloud With GPU Marketplace (2 minute read)

Nvidia's DGX Cloud Lepton is a new serv= ice designed to link artificial intelligence developers with Nvidia's netwo= rk of cloud providers, including CoreWeave, Lambda, and Crusoe. It provides= access to Nvidia's graphics processing units, making those chips more wide= ly accessible to all developers. The platform will have a marketplace of GP= U cloud vendors that developers can pick from to train and use their AI mod= els. Developers can choose which cloud provider to work with and even work = with multiple cloud providers.
=F0=9F= =9A=80

Science & Futuristic Technology

Robots Are Starting to Make= Decisions in the Operating Room (11 minute read)

Autonomous robots aren't being used to = operate on people yet, but the technology is almost capable of such a feat = and more autonomy is on the way. The technology could help hospitals meet t= he high demand for safety and consistency during surgeries, manage routine = tasks, and prevent mistakes. Future robots will use AI to autonomously mana= ge movements and decision-making during surgical tasks, reducing the need f= or constant human input. With the number of surgeons around the world decli= ning while the number of people who need surgery continues to increase, the= need for the technology is clear.
Valve CEO Gabe Newell's Neu= ralink competitor is expecting its first brain chip this year (3 minute rea= d)

Starfish Neuroscience, a brain-computer= interface startup incorporated by Valve co-founder and CEO Gabe Newell, pl= ans to produce its very first brain chip later this year. The company is cu= rrently seeking collaborators to open up avenues for the technology. Starfi= sh's goal is to create a smaller and less invasive implant than the competi= tion that can enable simultaneous access to multiple brain regions. Details= from the chip's current spec sheet are available in the article. The artic= le also links to a video of a talk from Valve that explains how the technol= ogy could enhance gaming.
=F0=9F= =92=BB

Programming, Design & Data Science

Stuff I learned at Carta (7= minute read)

Will Larson worked at Carta as its CTO = for the past two years. This post contains the most important lessons Larso= n learned in his time at the company, covering topics such as engineering s= trategy, large language model adoption, multi-dimensional tradeoffs, creati= ng software quality, and more.
OpenAI: Scaling PostgreSQL = to the Next Level (18 minute read)

Bohan Zhang, a member of OpenAI's Infra= structure team, recently shared the company's best practices with PostgreSQ= L at the PGConf.dev 2025 Global Developer Conference. The talk offers a gli= mpse into the database usage of one of the most prominent unicorn companies= and demonstrates how PostgreSQL can scale gracefully under massive read lo= ads. OpenAI utilizes managed databases on Azure, employing a classic Postgr= eSQL primary-replication architecture without sharding. This article covers= the content of Zhang's talk along with comments and includes some more det= ails about OpenAI's infrastructure.
=F0=9F= =8E=81

Miscellaneous

<= /div>
At Amazon, Some Coders Say = Their Jobs Have Begun to Resemble Warehouse Work (12 minute read)

As AI spreads through the labor force, = many white-collar workers say their work is becoming more routine, less tho= ughtful, and much faster-paced. Companies seem to be persuaded that AI can = increase productivity. Teams are getting smaller, but they're being expecte= d to output the same amount of work by using AI. Managers say that AI can r= elieve employees of tedious tasks and enable them to perform more interesti= ng work, but for inexperienced workers, the change resembles the shift from= artisanal work to factory work in the 19th and 20th centuries.
Ask HN: Go deep into AI/LLM= s or just use them as tools? (Hacker News Thread)

Large language models are becoming comm= odities. Many people use them, but few people are developing them from scra= tch. Developing large language models involves a lot of hardcore computer s= cience and math - people can use them without having to know any of the int= ernals. While it might be helpful to understand some of the internals and m= ath, it's probably best to stick to learning how to get good at using these= models and integrating what they do into agentic systems.
=E2=9A= =A1

Quick Links

Loving TLDR but want more A= I? (Sponsor)

We also write TLDR AI. Same format, all= about LLMs, GenAI, and deep learning. Delivered every weekday. Subscribe for free.
Root for Your Friends (4 mi= nute read)

Rooting for your friends is the best wa= y to improve your career, happiness, and the people you care about.
F2 (GitHub Repo)

F2 is a cross-platform command-line too= l for batch renaming files and directories quickly and safely.
Reinvent the Wheel (3 minut= e read)

You should learn from others, but you a= lso need to put your knowledge to the test so you learn enough about your f= ield to advance it.
On File Formats (3 minute r= ead)

This post presents things to consider i= f you are thinking about creating a new file format.
TrailBase (GitHub Repo)

TrailBase is an application server with= sub-millisecond latencies, eliminating the need for dedicated caches.
MCP is the coming of Web 2.= 0 2.0 (5 minute read)

MCP could inspire a new generation of d= evelopers to push for the web to go back towards its natural architecture, = programmable through janky specs that everybody adopted for the sheer joy o= f something fun to hack on.

Love TLDR? Tell your friends and get rewards!

Share your referral link below with friends to get free TLDR swag!
Track your referrals here.

Want to advertise in TLDR? =F0=9F=93=B0

If your company is interested in reaching an audience of tech executives, d= ecision-makers and engineers, you may want to advertise with us.

Want to work at TLDR? =F0=9F=92=BC

Apply here or send a friend's resume to jobs@tldr.tech and get $1k if we hire them!

If you have any comments or feedback, just respond to this email!

Thanks for reading,
Dan Ni &a= mp; Stephen = Flanders


Manage your subscriptions to our other newsletters on tech, st= artups, and programming. Or if TLDR isn't for you, please unsubscribe.
3D"" --kTz8PYjU-- --- þ Synchronet þ Vertrauen þ Home of Synchronet þ [vert/cvs/bbs].synchro.net .