- The following is a conversation all about the state-of-the-art in artificial intelligence, including some of the exciting technical breakthroughs and developments in AI that happened over the past year, and some of the interesting things we think might happen this upcoming year. At times, it does get super technical, but we do try to make sure that it remains accessible to folks outside the field without ever dumbing it down. It is a great honor and pleasure to be able to do this kind of episode with two ...

If you found it insightful, reshare it with your network.Find me → @_avichawlaEvery day, I share tutorials and insights on DS, ML, LLMs, and RAGs.Avi Chawla (@_avichawla):9 MCP, Agents, and RAG projects for AI engineers: https://t.co/fKTuaVMTc9 ...

If you found it insightful, reshare it with your network.Find me → @_avichawlaEvery day, I share tutorials and insights on DS, ML, LLMs, and RAGs. https://t.co/KpZaIUqSk2Avi Chawla (@_avichawla):Turn any Autoregressive LLM into a Diffusion LM.dLLM is a Python library that unifies the training & evaluation of diffusion language models.You can also use it to turn ANY autoregressive LM into a diffusion LM with minimal compute.100% open-source. https://t.co/sJGYiy009u ...

Core Viewpoint - Aswath Damodaran warns AI investors about the unsustainable valuation of companies in the sector, particularly highlighting a disconnect between infrastructure spending and future earnings [1][2]. Group 1: AI Market Concerns - Damodaran describes the current AI market as exhibiting signs of a "Big Market Delusion," where excessive optimism leads to unrealistic revenue expectations [2]. - He estimates that the AI industry must generate "two, three, four trillion in revenues eventually" to justify the current capital investments in Large Language Models (LLMs) [2]. Group 2: Company-Specific Analysis - Nvidia has been sold off by Damodaran due to its stock being "priced for perfection," indicating that it requires too many favorable conditions to break even [4]. - Despite Nvidia's strong position in AI infrastructure, its current valuation offers no margin for error, prompting a staggered exit from the stock over four years [4]. - In contrast, Microsoft is viewed as a more stable investment, with its cloud business seen as essential and more plausible in justifying its valuation [5]. Group 3: Performance Comparison - As of 2026, Microsoft shares have declined by 0.56% year-to-date, with a 8.24% drop over the last six months and 8.22% over the past year [7]. - Nvidia's stock is down 1.26% year-to-date but has seen a 5.50% increase over six months and a significant 57.46% rise over the year [7]. - Benzinga's Edge Stock Rankings indicate that Microsoft has a weaker price trend compared to Nvidia, which maintains a stronger price trend across all time frames [7][8].

Summary of Key Points from the Conference Call Industry and Company Overview - The discussion revolves around **Agentic AI** frameworks, which enhance traditional Large Language Models (LLMs) by integrating decision-making orchestrators and external tools, transforming them into autonomous problem solvers [2][4]. Core Insights and Arguments - **Agentic AI Workloads**: The paper profiles five representative agentic AI workloads: **Haystack RAG**, **Toolformer**, **ChemCrow**, **LangChain**, and **SWE-Agent**. These workloads are analyzed for latency, throughput, and energy metrics, highlighting the significant role of CPUs in these metrics compared to GPUs [3][10][20]. - **Latency Contributions**: Tool processing on CPUs can account for up to **90.6%** of total latency in agentic workloads, indicating a need for joint CPU-GPU optimization rather than focusing solely on GPU improvements [10][34]. - **Throughput Bottlenecks**: Throughput is bottlenecked by both CPU factors (coherence, synchronization, core over-subscription) and GPU factors (memory capacity and bandwidth). This dual limitation affects the performance of agentic AI systems [10][45]. - **Energy Consumption**: At large batch sizes, CPU dynamic energy consumption can reach up to **44%** of total dynamic energy, emphasizing the inefficiency of CPU parallelism compared to GPU [10][49]. Important but Overlooked Content - **Optimizations Proposed**: The paper introduces two key optimizations: 1. **CPU and GPU-Aware Micro-batching (CGAM)**: This method aims to improve performance by capping batch sizes and using micro-batching to optimize latency [11][50]. 2. **Mixed Agentic Workload Scheduling (MAWS)**: This approach adapts scheduling strategies for heterogeneous workloads, balancing CPU-heavy and LLM-heavy tasks to enhance overall efficiency [11][58]. - **Profiling Insights**: The profiling of agentic AI workloads reveals that tool processing, rather than LLM inference, is the primary contributor to latency, which is a critical insight for future optimizations [32][34]. - **Diverse Computational Patterns**: The selected workloads represent a variety of applications and computational strategies, showcasing the breadth of agentic AI systems and their real-world relevance [21][22]. Conclusion - The findings underscore the importance of a CPU-centric perspective in optimizing agentic AI frameworks, highlighting the need for comprehensive strategies that address both CPU and GPU limitations to enhance performance, efficiency, and scalability in AI applications [3][10][11].

What is Claude Code. Why is everyone so hyped about it. And what is it about this particular piece of software that versus what exists from OpenAI and Gemini and all this stuff.Like why is this captured everyone's imagination. If you really look at kind of what exists within Claude Code, you're calling out to a model and they gave it capability around sort of two big things. One is you can read and write files on your computer. And then two is that you can operate Unix, the bash commands that exist in your ...

As you uh develop more and more agents, as you take u these human-driven processes and agentify them, you have to think about like how do you bring that full comprehensive enterprise context to all all of these different agents. And ideally, we feel like AI should be like electricity. It just disappear.We don't, you know, be using AI without even knowing using AI. And it's almost like augumented reality before work. Arvin Ping, thank you so much for joining us for Greylock Change Agents.As you know, change ...

Core Insights - Jason Calacanis, a tech investor, believes Tesla's humanoid robot Optimus could surpass the company's automotive legacy, indicating a significant shift in focus for Tesla [1][3] Group 1: Visit to Tesla Optimus Lab - Calacanis visited Tesla's Optimus lab with CEO Elon Musk, observing the development of Optimus 3 and the engineering teams at work [2] - The visit took place two Sundays prior to the podcast episode recording, highlighting the ongoing advancements in Tesla's robotics efforts [2] Group 2: Future of Optimus - Calacanis predicts that Tesla will be remembered more for Optimus than for its cars, with Musk aiming to produce a billion units of the robot [3] - Musk has set a target price of $20,000 to $30,000 per unit for Optimus once mass production begins, indicating a strategic pricing approach [3] Group 3: Technology Integration - Large language models are expected to enhance Optimus robots, allowing them to understand and perform tasks that humans prefer to avoid [4] - Calacanis describes Optimus as potentially the most transformative technology product in human history, suggesting a future where there could be a one-to-one ratio of humans to Optimus robots [4] Group 4: Competition - Musk's Optimus project faces competition from Chinese Unitree robots, which have demonstrated impressive capabilities, including flips and dance routines [5]

All right. Thank you all for for joining for automating massive refactors with uh with parallel agents. Um super excited to talk to you all today about uh you know what we're doing with open hands to really automate large scale chunks of software engineering work.Lots of uh lots of toil related to tech debt, code maintenance, code modernization. Uh these are tasks that are super automatable. Uh you can throw agents at them, but they tend to be way too big for like you know a single just one shot.So it invol ...

Funding and Valuation - Anthropic has signed a term sheet for a $10 billion funding round at a $350 billion valuation [1] - Coatue and Singapore's sovereign wealth fund GIC are leading the financing [1] Company Background - Anthropic was founded in 2021 by former OpenAI research executives, including CEO Dario Amodei [2] - The company is known for developing a family of large language models called Claude [2] - Amazon has invested billions into Anthropic, while Microsoft and Nvidia announced plans to invest up to $5 billion and $10 billion, respectively [2] Competitive Landscape - Anthropic is competing with companies like Google and OpenAI, which has a valuation of $500 billion [3] - The company released three new models — Claude Sonnet 4.5, Claude Haiku 4.5, and Claude Opus 4.5 — late last year [3]