Supriyo SB Chatterjee

I attended the @CTHealth Connecticut Health Foundation 25th Anniversary Celebration at the @ctconventionctr Connecticut Convention Center in Hartford. I am a @CTHealth Health Leadership Fellow (Class of 2015). #HealthEquity

@sbc111 @nejmcatalyst @sbc111 Guten Tag, you can read it here: threadreaderapp.com/thread/2035053… Have a good day. 🤖

Agentic AI is advancing rapidly, and its use is growing. Here’s a helpful framework to learn Agentic AI. It’s a logical roadmap to build real skills, step by step. [ bookmark 🔖 this post for later ] 1. Agentic AI Introduction ➯ AI systems with autonomous decision-making abilities ➯ Main differences between intelligent agents and traditional AI ➯ Agent core functions: perception, reasoning, and action ➯ Business use cases in workflow automation 2. AI & ML Fundamentals ➯ Supervised and unsupervised learning approaches ➯ Neural networks and deep learning architectures ➯ Reinforcement learning powering autonomous agents ➯ Gradient descent and optimization methods for models 3. AI Programming & Frameworks ➯ Python libraries for creating AI agents ➯ API integration to enable function calls ➯ Frameworks: LangChain, AutoGen, CrewAI ➯ Data management and model orchestration patterns 4. Large Language Models (LLMs) ➯ Fundamentals of transformer-based architectures ➯ Tokenization and embedding methods for NLP ➯ Managing context window size and limitations ➯ Fine-tuning and advanced prompt strategies 5. Understanding AI Agents ➯ Types of agent architectures and design patterns ➯ Workflows for multi-agent collaboration and coordination ➯ Agent decision-making processes and reasoning chains ➯ Task-oriented vs. goal-oriented agent approaches 6. AI Knowledge and Memory Systems ➯ Managing short-term and long-term memory in AI ➯ Vector databases for knowledge storage and retrieval ➯ Implementing retrieval-augmented generation (RAG) ➯ Optimizing semantic search and document processing 7. AI Decision-Making & Planning ➯ Strategies for autonomous goal setting and execution ➯ Multi-agent coordination for problem-solving ➯ Hierarchical planning for intricate agent tasks ➯ Self-directed learning via feedback mechanisms 8. Advanced AI Learning & Adaptation ➯ Reinforcement learning with human feedback (RLHF) ➯ Dynamic optimization and control of prompts ➯ Instruction tuning for specific task performance ➯ Continuous agent improvement via reward training 9. AI Agent Deployment ➯ Cloud-based scaling of AI agent applications ➯ Model deployment using API architectures ➯ Performance tuning for low-latency responses ➯ Monitoring tools and maintenance protocols 10. Real-World AI Applications ➯ Automating business processes with intelligent agents ➯ Autonomous systems for research and data analysis ➯ Enhancing workflows through smart agent integration ➯ Decision-support tools for executive operations Agentic AI isn’t the next trend, it’s the next skill gap. Learn it to future-proof your career. 📌 Get Advanced ChatGPT Guide (free): bit.ly/3StIB3z 👉 Follow me @AndrewBolis for more and 🔄 Repost this to help others learn AI

@nejmcatalyst @threadreaderapp Unroll

Of about 51,000 MWB-enrolled members, about 78.5% of participants are engaging with BH services for the first time: nej.md/4rxKZFz

🧵 America is in the midst of a behavioral health (BH) crisis. Incidence of most BH conditions is increasing and affecting all age groups. In parallel, many areas lack an adequate supply of providers, exacerbated by high rates of burnout and low-efficacy case assignment models.

Most people don’t realise where the £180 million actually went at the A303 Stonehenge tunnel project. It wasn’t construction. It was years of: • Environmental impact assessments • Heritage & archaeological studies • Legal challenges and consultations • Design, engineering and traffic modelling • Public inquiries and revisions All before a single shovel hit the ground. This is the real problem in Britain: We don’t just waste money, we build systems that guarantee it. £180 million to build nothing.

Yesterday, Bessemer hosted our first-ever Robotics Day! We brought together over 80 founders, CEOs, researchers, and engineers for a deep dive into where physical AI is headed. We heard from the Bessemer team on our robotics roadmap, top researchers and builders from @DynaRobotics, xDOF, @perceptroninc, @CovariantAI, Bellboy, and Mind Robotics, and a live demo from @adrianmacneil at @foxglove. We closed the day with a fireside chat with @SebastianThrun, pioneer of autonomous driving and founder of @Google X and @Waymo . Thank you to everyone who joined us! The "ChatGPT moment" for robotics hasn't happened yet. But it's coming faster than most people think. Read our full roadmap: bvp.com/atlas/intellig…

10 Python Libraries Every #LLM Engineer Should Know @kdnuggets #TechHartford Interested in becoming an #AI LLM engineer? Here's a list of #Python libraries you'll find essential for your work. kdnuggets.com/10-python-libr…

5 Useful #Python Scripts for #SyntheticData Generation @kdnuggets #TechHartford Before you trust a library to generate your data, learn how to do it yourself and see where bias and errors actually begin. kdnuggets.com/5-useful-pytho…

What AI Startup Advisors See That Founders Often Miss @kdnuggets #TechHartford Understanding the gap between #startup aspirations and practical execution reveals important lessons about building sustainable #AI companies. kdnuggets.com/what-ai-startu…

@neo4j Thanks for a great post. This use case convinces me further that a graph model is a better paradigm than the traditional relational model.

Great #AI starts with great data. QIAGEN is transforming biomedical curation into an AI-ready graph intelligence platform with the Neo4j-powered Biomedical Knowledge Base (BKB) connecting 643 million curated relationships across genes, diseases, drugs, pathways, and omics data. The result? Life-saving medication for patients. bit.ly/4bzD1FN @QIAGENBiox @theohopkinson #GenAI #Graphdatabase

Innovating global regulatory frameworks for generative #AI in #medical devices is an urgent priority nature.com/articles/s4174… #MedTwitter #AImedicine #AI #HealthAI #AIinHealthcare #MedTech

#AI in #medicine shows promise but demands caution, expert warns itij.com/latest/news/ai… #MedTwitter #AImedicine #AI #HealthAI #AIinHealthcare #MedTech

Why you should not become an #AI expert @FastCompany fastcompany.com/91509048/why-y…

Is Google Lab's Stitch a Figma killer? f-st.co/J0JLXtz

Proceedings of the Fortieth AAAI Conference on Artificial Intelligence are now available to review online. The proceedings have been published in 48 consecutive issues which are all available here: aaai.org/proceeding/aaa…

Introducing Perplexity Health @perplexity_ai #AI #HealthAI perplexity.ai/hub/blog/intro…

NEW: Perplexity just launched a consumer-focused AI health tool Users can securely connect their health records and apps to Perplexity's AI answer engine via b.well’s national health data network. The health AI party just keeps getting bigger.

Introducing “vibe design” with Stitch @google Stitch is evolving into an #AI-native software design canvas that allows anyone to create, iterate and collaborate on high-fidelity UI from natural language. blog.google/innovation-and…