Benjamin Billings

I am excited to share the latest blog, co-authored by myself, Junhwi Kim and Vyas Sekar, detailing how we managed to successfully localized and improved LLM by combining Fine-Tuning and RAG. https://lnkd.in/gKDkgGPq

70

3 Comments

The Stoke Space team in Seattle has built a rocket company faster than many people can build software. Morgan Beller breaks down how this is helping them capture a huge opportunity: https://lnkd.in/gc5jv2-c 

23

2 Comments

🍜Anybody up for some ramen? I am excited to share that I am back in the garage, building again (it’s actually a house in the Berkeley Hills). We are taking the wraps off of Supercog AI, a new startup focused on solving Application Integration using LLM powered agents. My co-founder in this new venture is a great friend of mine (and a former co-founder partner), Alex Osborne. Alex and I have worked around the application integration space for a long time, and we are unrealistically excited about applying the GenAI stack to this problem. Not many people relish the task to get some bits to move from system A to system B. But it’s a critical job to be done, and that’s led to a lot of brittle scripts, a myriad of inflexible “no code” tools, and a lot of hand crafted SQL. But the power of the Large Language Model, trained on a huge corpus of information, offers a remarkable tool to solve this problem. The LLM you use today already knows the APIs of hundreds of popular systems. It knows the SQL dialects of every major database. It understands the semantics and data schemas of many popular SaaS systems. It understands much of the specific domain in which your business operates. I like to call the AI revolution “the last platform shift”. That may turn out to be hyperbole, but I definitely believe we are in the very early stages of seeing what this new stack can enable. I know we are hardly the first folks to claim that “amazing stuff is coming!”. But this is why Alex and I have decided to focus on a real and hard problem. If we can prove that LLMs can power a new way to solve this problem, it will be the first successful *new* approach in 15 years.. We will have much more to share about this new platform soon. In the meantime, if you’re interested in getting a peek - or even better if you have application or data integration tasks that you would like some help with - please reach out because we want to help. #genai #startups #backinthegarage

163

37 Comments

where’s OpenAI ‘s sama? oh look. he’s in elon’s shadow. makes sense. he’s shorter. Lol. the island of cannibals that is silicon valley has a new co-chair as mayor of sf. unelected. uninfluential in washington. has weasled his way into a transition team for sf. I wonder is watching elon w trump he found his trump locally in sf. sama is a mirror. he was for pg. he was also for many others. including elon. I guess this is a way to perhaps shore up political points locally in sf. to protect himself as best as he can. washington is a country wide, global game. kinda reminds me of another rebranding. im just jenny from the block. https://lnkd.in/e_jNXXRD if ai companies are to flourish OpenAI Google Microsoft Amazon Apple will require that startups are not somehow impeded. this is a game of davids and goliaths. where the slingshot appears innocuous. that is llms, transformers, access to privileged data. this out of no-where announcement demonstrates that this is a defense tactic by sama. he speaks fondly of sf and has done nothing until now. so his disingenuous nature finds yet another loop hole to worm into. we would be more supportive of his efforts except he’s created an industry for himself through a con game approach. why is he ceo of OpenAI when he’s there 6-7 hours a week. forgive me. isn’t a ceo supposed to be singularly focussed. his game is getting funding that invests in companies that he’s already invested in and as it’s a non for profit there are no conflicts of interest applied. wow. he uses other peoples money to invest in what he’s already invested in. sounds like y combinator showed him how to play the sv players. the next trick will be to do something no one has ever done. take a multi-billy non for profit and flip to for profit. as a non for profit there is no reporting required. there are tax incentives. a moral narrative to spin. then. oooops. how do we monetize this. we believe he will not succeed. to do so would cannibalize capitalism as it’s structured. it also would create a new pathway for universities which are by definition research centres. creating this precedent would alter their business. given that elon has beef with sama. and elon has never been more on top of his game. well. we find local political maneuvering is a way of squirming. OpenAI investors chiefly Microsoft are about to be thrown under the bus. unwittingly. recall it’s the talent at OpenAI that makes it special. the a team has left. the b team is in place. 2025 will not be good for OpenAI elon will have his revenge. 44 millys invested. for what. for sama to throw his weight around. he’s a lightweight. just look at him. lol. source: https://lnkd.in/e9WtQ48k < . >

2

What a week this has been for innovative companies in Seattle (first Smartsheet and now OctoAI). Today NVIDIA made public the news that they have acquired Madrona portfolio company OctoAI. Partnering with the founders since Day 1 (ok Day -1) has been one of the greatest experiences of my life. Luis Ceze, Jared Roesch, Tianqi Chen, Jason Knight, Thierry Moreau and Carlos Guestrin were all involved in the research and insights that helped create the TVM open source technology at the University of Washington computer science school and then the creation of OctoAI back in 2019 to drive commercial impact with this technology. From sorting out the terms of the original "pre-seed" round over a bottle of Caymus Vineyards wine to helping Amazon Web Services (AWS) understand the dynamics around the open source movement in Applied AI to learning our way in to product-market fit with the BEST inference optimization service available (OctoAI can help customers find their optimal fit of performance, customization, and cost), OctoAI has been the best! The company has helped set the table for AI-driven innovation for the years ahead and they will be a perfect compliment to the many capabilities at Nvidia. Luis, his leadership team, the board (including Carlos Guestrin Lip-Bu Tan and Mike Dauber) and so many "friends of OctoAI" know what a motivated, talented and authentic group of people are associated with this company. It is inspiring to see an acquisition at a total deal value of over $250 million in this highly dynamic and fast-moving phase of AI disruption. It is even more exciting to anticipate what will be possible with innovations powered by AI models and related technologies in the years ahead. Madrona has been investing in Applied AI for over a decade and we truly believe that we are just getting started! One of the hardest judgement calls as an investor and board member is to assess the benefits and risks of selling a company at a key inflection point in the journey. Nvidia made a compelling case that the impact for the OctoAI team and technology would be maximized within the market leader for AI infrastructure. I will be rooting every day that they all maximize this impact in the months and years ahead!

748

54 Comments

In the next few weeks, Snowflake and Databricks will host their annual summits - back-to-back - in San Francisco. The burden on the larger tech community is real: relationships will be tested. Dogs will go unwalked. Extended childcare will be arranged. Strict diets will be neglected, exercise foregone. Why are we doing this to ourselves? Because in 2024, these ecosystems are THE most important growth drivers. These data platforms largely dictate the fortunes of anyone in their orbit. No one is more trusted than the data platform. No technology is more central to more mission-critical workloads. As a result, no one is more influential in driving technology decisions. A primary data platform implies a primary ecosystem. And a primary ecosystem’s primary event is a big deal. At these events, seismic announcements will be made. Feature announcements that reshape reality for partners. Joint announcements with partners that drive critical awareness and pipeline. In parallel, decision makers will be exposed to new apps. New data providers. New changes in the SI landscape. New preferred SaaS tools for each workload (e.g. analytics, ETL, machine learning, dev ops, martech). For partners, these events are a pivotal time. They’ve spent the year building up the two critical aspects of ecosystem credibility: quality and volume of work. Now is the time to reap what has been sewn, and to plant for the forthcoming season. Year over year, outcomes from these events can and should improve. For those pursuing ecosystem-led growth - and, in 2024, it should be EVERYONE - navigating these events is an instructive microcosm for broader ecosystem participation: look for gaps within the platform’s ecosystem. Win with initial projects. Continue to establish yourself in the ecosystem with more marquis projects. Promote successes effectively within the ecosystem. And, importantly, don't bite off more than you can chew. Be disciplined and measured. Credibility within the ecosystem is built over years, and can be lost in a day. Personally, I’m very much looking forward to seeing all of you at both of them. Maybe even at the dingy hotel gym this year. I swear I’m gonna make it this time… #snowflake #databricks #ecosystem #data

55

7 Comments

Unfortunately OpenAI o1 isn't "reasoning" or "thinking" as we humans define it. OpenAI has (intentionally?) muddied the UI/UX waters by introducing verbs like "thinking" / "analysing" etc. to encourage users to anthropomorphize what is going on behind the scenes. o1 is more like iterative prompt generation, and then doing next token prediction for each prompt using an agent. So, yes — it may generate better answers in some contexts, especially the ones where RLHF has been used on its chain-of-though processes (prompt generation), like maths and programming, but its likely to fall into many of the same general traps as next token prediction. The structure and sequence of tokens doesn't *generally* allow you to infer logic. Syntax is not semantics, and the jury is still out on what level of semantic generalisation LLMs can encode, to say nothing of their implementations of algorithms for logical inference and search. #LLM #OpenAI #o1

73

27 Comments

Recently took on the challenge of building an app from scratch using a new tech stack by leaning into ChatGPT. The outcome was much better than I expected and I only had to deep dive into the code in a few tricky situations. https://www.jimoto.life/ (It originally started out as an app to help my in-laws find a good neighborhood to move to, but I got carried away with a few other fun features... 😅 code, layout, even the background image and color schemes are AI generated/suggested.) ChatGPT is a must have for anyone in tech for sure . #ChatGPT #AI #AppDevelopment

5

1 Comment

If we want to see more diverse top companies like NVIDIA, we need to pave the way at the earliest stages—venture capital is that on-ramp. It was a pleasure to join Emerging Manager Office Hours hosted by Venture Forward earlier this week. Thank you to Rhianon Anderson and Maryam Haque for organizing such a value add event! A few (other) big takeaways to share: 👉 Distinguish the Curious from the Serious: Focus your time on LPs who are truly engaged. There are many LPs out there still excited about and committed to the VC asset class. 👉 Showcase Portfolio Performance: With liquidity tight, highlight KPIs, revenue growth, and the value your companies are building now to position for success when markets rebound. 👉 Clearly Differentiate: Provide concrete examples of how your approach stands out and drives unique outcomes, i.e., case studies. It was a pleasure connecting with friends and meeting inspiring next-gen GPs shaping the future of venture capital: William Crowder (Aperture® Venture Capital), Joshua Weed (Veteran Ventures Capital), Sonam Velani (Streetlife Ventures), Marcos Gonzalez (VamosVentures), Erica Duignan Minnihan (Reign Ventures), and Jenny Xiao (Leonis Capital). Here’s to collaboration, innovation, and creating lasting impact together.

63

5 Comments

TL;DR Impacted by the Tesla layoffs. A wild ride worth reflecting on and taking some time to recharge the batteries. It all started innocently enough with a LI message from a Tesla recruiter in the early months of 2020, and little did I know the entire world would be turned upside down for the better part of 2 years. What was supposed to be an onsite interview suddenly became a virtual interview, and ended up as a WFH onboarding. Tasked with building a scalable Learning platform from scratch for our almost 160k+ strong global workforce of employees and partners as we migrated systems with an early 2021 delivery deadline seemed an almost insurmountable ask at the time, very much complicated by being a new hire and trying to get things done in a virtual work environment the entire company was figuring out. Against this backdrop, I got to witness first-hand the undeniable dedication, efficiency, and passion of all of my colleagues at Tesla. There was no hesitation when we had to spend late nights until 3am during Thanksgiving week nor the countless weekends collaborating to ensure we got the job done. And got the job done we did. Fast forward 4 years, and together we’ve built an enterprise Training platform that has scaled to support Tesla’s almost 300k global workforce of employees and partners through eLearning, In-Person instruction, and on-the-job Assessments. We had the unique opportunity of supporting all parts of our business globally from HR to Engineering to EHS&S to Energy to Manufacturing to SS&D and to Supply Chain. While it’s certainly a disappointment for myself as well as for my many dedicated and passionate colleagues who have been impacted, I take immense pride in knowing that we all contributed to a greater cause and mission than ourselves: Accelerating the world’s transition to sustainable energy. And even though we’ll no longer carry the torch for now, I know that through our many contributions that we’ve left Tesla in a better place than when we joined, and Tesla is absolutely primed for the next growth phase. We’ll all be cheering from the sidelines! I would be remiss if I did not mention some of the talented colleagues who were instrumental to my Tesla journey: Malia Schober Eben Carey Dave Parrent Shelby Callahan Avinash Matha Janelle Flores Lixin Wang Prateek Khanna Pankaj Saini Ghizlane Bounoua Aarthi Palaniswamy Renee Hu Lizette B., PsyD Molly Caddam David Dodge, PMP Dillon Lee Matt Primeau Jack Sylvester Kevin McMurphy Alfredo Aguirre Stephanie Hernandez Divya H Manasvi Komma Mike Nobles Carly Bermensolo Rafae Shaik Xiechuan Liu #Tesla #Teslalayoffs #TSLA

477

37 Comments

To all my fellow Arizonans in particular… Here’s the Cliff Notes version of why our Silicon State focus on “atoms” in addition to bits is SO on point for our times. The world of atoms shall be underserved no longer. And Avanced Manufacturing (of all kinds of things in all kinds of industries) is right on the $! 🙏🏼🌵🚀 https://lnkd.in/gNbGmPsC

11

1 Comment

Dear friends, and colleagues, we are starting a new STEM program for youth girls, and wondering if you would help support the its goals, by donating a bit through your organization match, or directly via PayPal. My 14 yr old daughter Elle and her friend participated in a Google sponsored Girls in Tech robotics event, and through that they both felt what it is like to engage in technology in a fun and challenging robotics competition, in which they came in FIRST place. They never before felt such empowerment and satisfaction through their efforts! Purely due to that experience they are both fired up to do more. X Academy is sponsoring this program. This program aims to inspire and empower the next generation of female innovators. This new girls team will compete using the VEX V5 robotics system, and this is where we need help sourcing the funds for the equipment. Donate Now: PayPal Donation Link https://lnkd.in/gW7HqRAP Thanks for your support, and full details below, -Russsell What is X Academy The X Academy (https://www.xacademy.org/) is a parent-led 501c3 nonprofit celebrating its 10th year of providing STEM enrichment for middle and high school students throughout Santa Cruz County. Over 500 middle school students experienced the joy of challenging math through the Math Circle and Julia Robinson Math Festivals. In June this year, the X Academy high school robotics team won 3rd place at the MATE ROV World Championship. Why Your Support Matters STEM education (Science, Technology, Engineering, and Mathematics) plays a crucial role in shaping our future.By supporting our girls’ robotics team, you are helping to build a strong foundation for these young women to excel in these fields and contribute to groundbreaking advancements in technology and engineering. We're passionate about ensuring that robotics is a welcoming and empowering space for all. We want girls to feel like they have a place in robotics. We recognize that historical biases, stereotypes and limiting self-perceptions can discourage girls from pursuing STEM fields, and we're committed to changing that narrative. How You Can Help To make this initiative a success, we need to raise a fund of $7500 for the VEX V5 robot (parts and tools), playfield setup, and competition fees. Your generous donation will directly support these essential components, ensuring that our team has the resources they need to compete and thrive. Corporate Matching Opportunities If you are affiliated with a company that participates in employee donation matching programs, we encourage you to donate through your company’s Benevity portal. Many companies offer to match donations, effectively doubling your impact. X Academy is listed on Benevity, making it easy for your company to contribute. We are incredibly grateful for your support and commitment to fostering the next generation of female leaders in STEM. Together, we can make a significant difference.

8

𝗧𝗵𝗶𝗻𝗸𝗶𝗻𝗴 𝗶𝘀 𝘄𝗿𝗶𝘁𝗶𝗻𝗴. Paul Graham's latest essay hits hard on this point: AI is rapidly turning writing into an optional skill, much like how industrialization made physical strength less of a daily necessity. In the future, 𝘵𝘩𝘪𝘯𝘬𝘪𝘯𝘨—the kind that requires clarity, depth, and precision—will be something we actively choose to exercise, just as we work out today to stay fit. The result? A world divided into "thinks" and "think-nots." Those who choose to write will sharpen their thinking. Those who don’t may lose touch with the unique clarity that comes from grappling with words on a page. We’re at an inflection point: do we let AI take the wheel, or do we commit to keeping our mental muscles strong? https://lnkd.in/gKvgVT3Y

1

1 Comment

"Finding community is so much harder than founding a company 🚀" That's what a friend said the other day, who just moved to SF. I wished I could have pointed him to something like a magic list of all the SF-based communities directed at new founders. Before moving to SF myself to build my current startup, I wanted such list as well. In fact, I spent a good amount of days googling and googling "top hackerhouses in SF" and "founders community SF", only to find few results, things I already knew and outdated blogposts. Thankfully, On Deck's SEO was solid, so I applied and got accepted and that was my lucky baptism in SF's founder communities. Some folks like Garry Tan, say that SF's uniqueness is not its taxes, but its people. It's the builders, the hustlers, the AI startuppers, even -maybe- the VCs. That's definitely true, as long as you're able to find them! If you can't find them in the first place, there's no point in paying the SF rent, nor a reason to move there. You may as well live in Treasure Island! (fun fact, I do) So, for those like my friend, like myself, and for you, here's *our* imperfect list. Shoutout to the tons of friends who invited me to, hosted me at and co-infiltrated different communities and events over these past few months. Did I miss any other communities/accelerators/hackerhouses? Definitely comment below or DM me and I'll edit ASAP. On Deck Julian Weisser Noémie Federico South Park CommonsZ Fellows Inception Studio Entrepreneur First Founders, Inc. A16Z GAMES SPEEDRUN Y Combinator Techstars Unshackled Ventures Sequoia Capital Pear VC Alchemist Accelerator 500 Global Race Capital HF0 AGI House Rocky Yu SHACK15 Ventures SHACK15 Studio 45 Nate Padgett The Battery The Commons The Long Now Foundation Commonwealth Club World Affairs

95

12 Comments

Terence Tao on LLM There is too much hype, hucksters and dumb money poured into GenAI and LLM now. It’s refreshing to hear from expert voice for once. Terence Tao is a UCLA math professor, Fields Medal winner and is considered the world’s greatest living mathematician. I overlapped a year with him at UCLA in 1996 and I met him when he was TA for a Linear Algebra course I took my senior year. Terence recently gave his honest feedback on GPT-o1 here https://lnkd.in/gq_6bcwM, and was subsequently interviewed by The Atlantic https://lnkd.in/gv6fgqrB I will summarize this interview: 1. OpenAI o1 is a mediocre or incompetent research assistant. 2. It does not “reason”, is not a source of knowledge or ideas, but is a really useful glue. 3. We will always need humans and AI. They have complementary strengths. AI is very good at converting billions of pieces of data into one good answer. Humans are good at taking 10 observations and making really inspired guesses. I cannot agree more with the “Mozart of Math”. Especially on point number #2. LLM is very good at doing relational joins of non-relational unstructured data. Once you realize its true strength and what it isn't, LLM can be a very powerful tool. To read more of my ponderings on "AI" and other tech stuff, subscribe to my substack: https://lnkd.in/gtknApfw

43

9 Comments

BREAKING: Aptos sets a new record on 8/14/24 with 326 million user transactions in a day (see https://lnkd.in/gbnfMGmx). That's an average of ~3766 user txns per second over a day, with fees as low as ~$0.00005 per txn. The network continued normal operations during the day without issues. This achievement more than doubles the previous record set (also by Aptos) of 154.4 million on 5/25/24 (~1787 txns per second over that day). The previous records were 95.6 million (Aptos) on 5/24/24 and another chain with 65.8 million in 7/23. Aptos now holds the top three records for daily transactions and is the clear infrastructure leader in mainnet throughput and speed in production. Remarkably, these transactions are not simple or trivially parallelizable either. They access global state (shared counters) that serialize transactions on other networks and are only parallelizable on Aptos. Web3 infrastructure is ready for the large-scale and novel use cases in the world. Major innovation to continue rolling out in the coming months!

338

18 Comments