We’re happy to announce that we have raised a $2.8M Pre-Seed round! The round includes LongHash Ventures, Inception Capital, Protocol Labs, TRGC, L2 Iterative Venture (L2IV), Zero Knowledge Ventures, Fenbushi Capital, Quaker Capital Investments, LLC, Orange DAO, and Escape Velocity (EV3)! We’re building the world’s most advanced AI infrastructure company, starting with an integrated hardware and software compute protocol to train and fine-tune AI models. Looking forward to building great things! Thank you everyone for believing in us! If you’re interested in learning more about Lumino, go to: https://lnkd.in/gfgfd9c4 If you’re interested in learning more about career opportunities, head over to https://lnkd.in/g-2z2qse
About us
Decentralized compute protocol to train and fine-tune AI models
- Website
-
https://www.luminolabs.ai/
External link for Lumino
- Industry
- Technology, Information and Internet
- Company size
- 2-10 employees
- Headquarters
- San Francisco Bay Area
- Type
- Privately Held
Locations
-
Primary
Get directions
San Francisco Bay Area, US
Employees at Lumino
Updates
-
Come help us build the world's most advanced ML training platform! If you're looking for a role as a Machine Learning Engineer, apply here: https://lnkd.in/gggFR8x8 The role is a combination of applied and research ML. You'll be working on training and fine-tuning pipelines for LLMs and deep learning, building out predictive models for our product, and researching new ways to do distributed training!
Machine Learning Engineer
jobs.ashbyhq.com
-
We're #hiring a new Protocol Engineer in San Mateo, California. Apply today or share this post with your network.
-
Lumino reposted this
Thank you Grant Long for a great chat! Training AI models requires large amounts of computing power, which has led to a price surge for GPUs. Additionally, large enterprises have reserved GPUs and have made it impossible for others to rent out GPUs. Finally, large public cloud providers are able to pick and choose who has the ability to build AI, potentially leading to censorship or concentrating power in the few. Yogesh Darji and I are building a decentralized AI training platform that incentivizes independent data centers and GPU owners to contribute capacity to fight these problems. We're hiring great engineers to help us build the future of AI. If you're interested in learning more, DM me or visit: https://lnkd.in/gNsz9C5i
If AI is as fundamental as many of us believe... Can humanity afford for a few mega corporations to control compute? Eshan Chordia sat down with me to discuss his new company Lumino, which strives to help us avoid a centralized future for AI. Lumino is building an AI training and compute cloud that incentivizes independent data center and GPU owners to contribute their extra capacity, with cryptoeconomic incentives. After a fascinating convo, it is clear to me that Eshan is the right man to take on this problem. Wishing him and the company the best. https://lnkd.in/geQmEGsM
Resisting Big Tech with decentralized AI compute: Lumino AI founder Eshan Chordia
https://www.youtube.com/
-
Lumino reposted this
S2 E3 of the DFI Podcast with Eshan Chordia and Yogesh Darji is live! https://lnkd.in/g3z6i2sR In this episode, we celebrate the launch of Lumino AI 🙌 🎉 For any Engineers looking to join a startup in the Decentralized AI space with a stellar set of co-founders - please check out https://www.luminolabs.ai/
S2E3: Lumino Labs Announcement w/co-founders Yogesh and Eshan, and the Future of Decentralized AI
https://www.youtube.com/
-
Lumino reposted this
We recently participated in Lumino's latest pre-seed round, and we couldn't be more thrilled to back Eshan Chordia & Yogesh Darji's vision for Lumino AI's decentralized compute protocol that leverages idle GPU resources to offer cost-effective & accessible AI training. The demand for GPUs to power the $600B AI training market is skyrocketing, yet centralized providers struggle to keep up. Lumino aggregates supply across diverse Compute Providers to enable up to 80% savings vs centralized competitors. We believe their orchestration tech & dev platform will be key to mainstream adoption. Please find our detailed report on our thesis behind investing in Lumino AI. Kevin Lee, CFA Dennis J.H. Won Arhat Bhagwatkar Weikeng Chen Kony Kwong https://lnkd.in/gWeVDPSb
Why We Invested in Lumino AI
l2ivresearch.substack.com
-
Lumino reposted this
We’re building exactly this at Lumino AI. Serverless AI compute platform that gets you to production in minutes and is much cheaper than the large public cloud companies!
During the mid 2010s, I made a supposition that most unprofitable VC backed companies were spending $.40 of every $1 raised on FB and Google ads and AWS compute. It turned out to be largely right. Unfortunately, we are back to this same cycle in AI with NVDA, but I worry that it’s now $0.60-75 of every $1. Every entrepreneur should be aggressively trying to find a way to shrink their compute OPEX and use a managed service that delivers production quality speed. Otherwise this is all just sandboxing, toy apps and means little of anything. Boards should be asking for OPEX breakdowns and pushing their companies to look at every possible way to spend as little as possible for compute.
-
Lumino reposted this
We’re building exactly this at Lumino AI. Serverless AI compute platform that gets you to production in minutes and is much cheaper than the large public cloud companies!
During the mid 2010s, I made a supposition that most unprofitable VC backed companies were spending $.40 of every $1 raised on FB and Google ads and AWS compute. It turned out to be largely right. Unfortunately, we are back to this same cycle in AI with NVDA, but I worry that it’s now $0.60-75 of every $1. Every entrepreneur should be aggressively trying to find a way to shrink their compute OPEX and use a managed service that delivers production quality speed. Otherwise this is all just sandboxing, toy apps and means little of anything. Boards should be asking for OPEX breakdowns and pushing their companies to look at every possible way to spend as little as possible for compute.