MJ
@pythonym
Followers
177
Following
56K
Media
982
Statuses
8K
Write about AI/ML tools, and concepts!
Check out 👉
Joined November 2014
Learning LLMs in 2025! - Shift from "Just Using" to "Building With": The barrier to using LLMs via chat interfaces is zero. Value is now in building robust, reliable, and innovative applications on top of them. - Multi-Modality is the Default: Text-only models are still
2
0
0
🔥 Mastra It is a framework for building AI-powered applications and agents with a modern TypeScript stack. Why Mastra? Purpose-built for TypeScript and designed around established AI patterns, it gives you everything you need to build great AI applications out-of-the-box.
0
0
0
Kudos, Nikhil Naz, for exposing the farce that was Lionel Messi's India tour. The cost of this tour was Rs. 120 crore. At the same time, India's national football league stands suspended because nobody is willing to shell out Rs. 35 crore.
183
3K
9K
🔥 Why Causal Attention? When training a generative LLM, there’s one hard rule: The model cannot see the future tokens. That’s enforced using causal attention, also called masked attention. It ensures that, at any moment, the model can only look at the current token and
0
0
1
🔥 Astron It is an enterprise-grade, commercial-friendly Agentic Workflow development platform that integrates AI workflow orchestration, model management, AI and MCP tool integration, RPA automation, and team collaboration features. Key Features Enterprise-Grade High
0
0
0
âš¡ A Deep Dive into Probabilistic Models Probabilistic Models serve as the backbone of learning in the realm of machine learning. It helps in decoding the hidden patterns within the data, thus enabling us to make informed predictions about unseen data. It truly excels at
0
0
0
In this video, we take Retrieval-Augmented Generation (RAG) to the next level using OpenSearch, combining Hybrid Search (Keyword + Semantic Search) with Docling-based document filtering for cleaner, smarter, and more accurate retrieval. GitHub Notebook:
github.com
Understanding and implementing NLP and CV libraries with ML standpoint. - Mayurji/Explore-Libraries
0
0
0
The law was upheld!
Pune Porsche case: Bombay HC denies bail to the builder father and six others. 1) Dad orchestrated the entire cover-up. 2) Mom walked into hospital and gave her own blood to replace her son's sample. 3) Doctors took ₹3 lakhs to swap the vials, forge the MLC register, and issue
0
0
0
🔥 PipeCat It is an open-source Python framework for building real-time voice and multimodal conversational agents. Orchestrate audio and video, AI services, different transports, and conversation pipelines effortlessly. Why Pipecat? - Voice-first: Integrates speech
3
0
0
I am starting to feel the Indian economy is Dead; fake GDP numbers cannot revive it. #stockmarketsindia #USDINR
0
0
0
🔥 OpenAgents It enables general users to interact with agent functionalities through a web UI optimized for swift responses and common failures. It offers developers and researchers a seamless deployment experience on local setups, providing a foundation for crafting
0
0
0
It's the end of 2025, Modi still thinks that changing the name of a policy is still an achievement.
0
0
0
0
0
0
🔥 Key Emerging Themes from NeurIPS 2025 - Large Language and Foundation Models - RL and Agents - Datasets, benchmarks, and Evaluation. - Applications and Interdisciplinary AI. #NeurIPS2025 Read here for more:
medium.com
Explore the trends that the field of AI is riding on!
0
0
0
âš¡ VoltAgent It is an open-source TypeScript framework for building and orchestrating AI agents. You can build production-ready agents with memory, workflows, tools, and built-in LLM observability. Why VoltAgent? - Production-Ready from Day One: Ship agents with built-in
1
1
4
🔥 Notable Paper Examples from NeurIPS 2025 Gated Attention in LLMs (Best Paper) – It introduced a simple yet powerful modification to Transformer attention: inserting a head-specific sigmoid gate after the standard softmax operation. Across experiments on 1.7B and 15B models
intuitionlabs.ai
An educational overview of the NeurIPS 2025 conference. Learn about key trends in AI research, including LLMs, major awards, acceptance rates, and new paper tra
0
0
0
🔥 AgentGPT It helps you to assemble, configure, and deploy autonomous AI Agent(s) in your browser. 🤖 You can name your own custom AI and have it embark on any goal imaginable. It will attempt to reach the goal by thinking of tasks to do, executing them, and learning from the
0
0
0
🔥 XAgent It is an open-source experimental Large Language Model (LLM) driven autonomous agent that can automatically solve various tasks. It is designed to be a general-purpose agent that can be applied to a wide range of tasks. It is designed with the following features:
0
0
0
🔥Deep Dive into Tree-based ML Model It is the most common supervised learning method, built with two entities: nodes and branches. The tree structure is formed by recursively splitting a training sample, using different features from the dataset at each node that splits the
1
0
2