TuringPost @TheTuringPost

Newsletter exploring AI&ML - AI 101, Agentic Workflow, Business insights. From ML history to AI trends. Led by @kseniase_ Know what you are talking about👇🏼 turingpost.com/subscribe Join over 95,000 readers Joined June 2020

Tweets

17K
Followers

76K
Following

13K
Likes

12K

TuringPost @TheTuringPost

a hour ago

ModelOpt, @nvidia TensorRT Model Optimizer – open-source toolkit to accelerate models • Supports end-to-end optimizations: quantization, pruning, distillation, speculative decoding, sparsity • Works with Hugging Face, PyTorch, ONNX models • Integrates with NeMo, Megatron-LM,…

0 7 31 1K 13

Download Image

TuringPost @TheTuringPost

a day ago

If you want the fastest way to build Elixir apps in-browser – try Phoenix.new → phoenix.new It lets you: • Spin up real Elixir apps in the browser – no setup, no yak-shaving • Run tests with full root access • Interact with the UI in a headless…

0 1 19 4K 5

Download Image

TuringPost @TheTuringPost

2 hours ago

Hardware becomes 100x more efficient. We see this with CPUs, and now we should see it with GPUs. @lqiao, co-founder & CEO of @FireworksAI_HQ and former head of PyTorch at Meta More about inference, alignment, and the future of AI infrastructure: youtube.com/watch?v=bb4FHF…

0 1 6 1K 3

Download Video

TuringPost @TheTuringPost

a day ago

One of the most comprehensive survey on Agentic RL (reinforcement learning) Covers: - From passive LLMs ➝ active decision-makers - Key skills: planning, tools, memory, reasoning, reflection, perception - Uses: search, code, math, GUIs, embodied agents - Benchmarks,…

7 84 424 27K 506

Download Image

TuringPost @TheTuringPost

a day ago

A new way for LLM agents to adapt ➡️ Memento uses memory + reinforcement learning (RL) to perform case-based continual learning for LLM agents. It has a Case Bank – a growing memory of past experiences, that updates during use, to help agents learn on the fly. This setup is…

6 27 126 8K 102

Download Image

TuringPost @TheTuringPost

2 days ago

Protecting AI Agents with AI A new paradigm in secure AI → fine-tuning SLMs for сontext-aware data protection that identifies and stops sensitive data leaks. A great opportunity to explore how it works is @predibase upcoming Webinar: bit.ly/4m7xMAz Date: September…

0 3 23 5K 4

Download Image

TuringPost @TheTuringPost

2 days ago

HeteroScale – a coordinated autoscaling framework from @ByteDanceOSS for LLMs with the Prefill–Decode (P/D) architecture > It improves GPU usage by 26.6% and saves hundreds of thousands of GPU-hours daily! How does HeteroScale achieve this? - It balances prefill and decode…

1 3 17 5K 15

Download Image

TuringPost @TheTuringPost

4 months ago

7 free resources to master Multi-Agent Systems (MAS): ▪️ CrewAI tutorials ▪️ Building with CAMEL multi-agent framework ▪️ Lang Chain multi-agent tutorial ▪️ "Multiagent Systems: Algorithmic, Game-Theoretic, and Logical Foundations" book Courses ▪️ AI Agents: From Prompts to…