California's Senate Bill 1047 is deeply flawed. Holding developers liable for misuse of their models won't improve AI safety, but it will negatively impact innovation. Like many other startups, our technology is possible because of access to open source LLMs. Companies of every size, from enterprises to start-ups, benefit from open source innovation. Sharon Zhou, PhD recently spoke on a panel with Matt W., GM of AI at The Linux Foundation, Charles Xie, CEO and founder of Zilliz and creator of Milvus, and Christopher Cuong T. Nguyen, CEO and co-founder AITOMATIC about the importance of an open ecosystem and implications of SB 1047. Thanks to The AI Alliance and Cerebral Valley for bringing more awareness to SB 1047!
Lamini
Software Development
Menlo Park, California 6,282 followers
The LLM Development Platform for Enterprises
About us
Lamini makes it possible for enterprises to turn proprietary data into the next generation of LLM capabilities, by offering a platform for in-house software teams to uplevel to OpenAI-level AI teams and to build within the security of their existing infrastructure.
- Website
-
https://lamini.ai
External link for Lamini
- Industry
- Software Development
- Company size
- 11-50 employees
- Headquarters
- Menlo Park, California
- Type
- Privately Held
- Founded
- 2022
Locations
-
Primary
Santa Cruz Ave
Menlo Park, California 94025, US
Employees at Lamini
Updates
-
Lamini reposted this
The value of #AI is significantly enhanced when it can operate in real time. Learn about reducing latency, managing costs, and achieving real-time access, as discussed by Lamini CEO Sharon Zhou, PhD at the Real-Time Data Summit, and how #Aerospike has been at the forefront of powering real-time AI applications. https://lnkd.in/gYqg6ZNs
-
Lamini reposted this
New course on DeepLearning.AI: Improving Accuracy of LLM Applications. More details ➡️ https://go.fb.me/zfwvd8 Created in collaboration with DeepLearning.AI, Meta and Lamini, this free course covers: • Building a text to SQL agent with Llama 3. • Evaluation frameworks. • Instruction & memory fine-tuning. • Breaking fine-tuning myths. • PEFT techniques like LoRA and MoME. • Processes for generating training data. • More
-
-
We're #hiring a new Developer Advocate Lead/Specialist in Menlo Park, California. Apply today or share this post with your network.
-
🔥 How did we scale our GPU cloud service so quickly? We worked with Supermicro to bring their high-performance GPU servers with AMD Instinct™ MI300X Accelerators to our data center, of course!
🎉We're excited to share how Lamini, a leading AI cloud service provider, is leveraging our Supermicro GPU Servers to support their large-scale LLM tuning and generative AI models. With our building block architecture, Lamini was able to quickly bring up their AI cloud service. 🔗Read more about this in our Success Story: https://hubs.la/Q02LtCfc0 #Supermicro #GenAI #AI #Lamini #LLM #innovation #datacenter
-
-
We're #hiring a new Sales Engineer (West Coast) in Menlo Park, California. Apply today or share this post with your network.
-
🎉 🎉 🎉 In collaboration with DeepLearning.AI, Andrew Ng and Amit Sangani, we've launched our new short course, Improving Accuracy of LLM Applications. You'll learn all the latest techniques for fine-tuning LLMs to achieve high accuracy and reliability.
Struggling with inconsistent outcomes in your LLM applications? Our new short course, Improving Accuracy of LLM Applications, is here to help! Developed in collaboration with Lamini and AI at Meta, and taught by industry leaders Sharon Zhou, PhD of Lamini and Amit Sangani of Meta, this course provides a structured approach to improving your LLM apps. 🛠️ What you'll do: - Build a text-to-SQL agent using Llama open source models - Implement prompt engineering and self-reflection to enhance model behavior - Fine-tune models with techniques like LoRA and memory tuning to reduce hallucinations - Build an evaluation framework to systematically measure performance 👉 Start today: https://hubs.la/Q02LgZyg0
-
How do you distill the knowledge of a large model into a highly specialized and efficient model to achieve 100ms latency? We're glad you asked. Lamini Memory Tuning reduces hallucinations by 10x over traditional RAG or fine-tuning by: 📍 Selective information retention 💎 Compact representation ⏱ Efficient retrieval Learn more in our latest blog post. https://lnkd.in/gA9rgWpP
Under 100ms Latency with Use Case Specific Knowledge Distillation | Lamini - Enterprise LLM Platform
lamini.ai
-
Lamini reposted this
The amazing Sharon Zhou, PhD CEO and Co-Founder of Lamini talked about open source LLMs and GenAI. Lamini was built from the ground up with enterprises in mind and is focused on delivering high generative AI accuracy and scalability. #Innovation #KPMGTechandInnovation24 #Lamini KPMG US You missed a great event Mike Dauber !!
-
-
We're honored to be featured as a Rising Star ⭐️ in the Forbes 2024 Cloud 100! Thankful for the recognition Forbes Bessemer Venture Partners Salesforce Ventures. https://lnkd.in/gWnvMWCN #Cloud100 #RisingStar
Cloud 100 Rising Stars 2024: The Next Generation Of Superstar Cloud Companies
social-www.forbes.com