Priyanka Nath

Something I've written on how a T20 batting innings can be assessed. This framework can be extended to assessment of individual players as well, which I will do in the susequent posts. For now, please feel free to comment and if you like the content, please subscribe. #cricket #cricketanalytics #data #dashboards #datavisualization #sportsanalytics https://lnkd.in/gb7T-UzJ

28

6 Comments

Check out this highly recommended course for those interested in optimizing their models: - Introduction to Quantization - Symmetric and Asymmetric Quantization Modes - Hands-on Implementation from Scratch using PyTorch - covering Per Tensor, Per Channel (Axis), and Per Group techniques - Weights Packing: Exploring 2 or 4 Bits Quantization Methods https://lnkd.in/gMAuWXHm

43

Hello Everyone Your learning - #learnwithvanama ---- 🔍 Exploring Best Practices in Retrieval-Augmented Generation (RAG) 📄 I recently delved into the insightful paper "Searching for Best Practices in Retrieval-Augmented Generation" by Xiaohua Wang et al., which offers a comprehensive survey on RAG techniques. This innovative approach merges large language models with information retrieval, enhancing content accuracy and relevance by tapping into external sources. Key Findings: - Mitigating Hallucinations: RAG proves effective in reducing inaccuracies and improving response quality, especially in specialized fields. - Multimodal Enhancements: Techniques for multimodal retrieval enhance question-answering related to visual inputs and expedite the generation of multimodal content. - Challenges: - Implementation complexities and extended response times are notable hurdles. - Integration Potential: Combining RAG with reinforcement learning or few-shot learning could yield more robust AI systems. Best Practices: - Continuous Evaluation: Regularly assess outputs to identify areas for enhancement. - Transparency: Clearly explain how outputs are generated to build user trust. - Personalization: Use customer integration data to tailor AI features effectively. The paper serves as a significant step forward in refining retrieval-augmented generation. While challenges persist, the proposed strategies and best practices offer a solid foundation for optimizing RAG and advancing AI capabilities. Embracing these insights can drive the development of more knowledgeable and contextually-aware AI systems. Link : https://lnkd.in/e3YJfq4q #AI #MachineLearning #Research #DataScience #Innovation #TechTrends

22

As we celebrate the 7th anniversary of Observe.AI, I find myself reflecting on how far we've come. What started as a simple idea in 2017 has grown into the leading AI company in one of the world’s oldest and largest markets—Contact Centers. I remember visiting Manila, eager to learn about this industry, and being struck by the outdated, manual workflows agents, managers, and supervisors were struggling with. I saw agents juggling 2+ monitors, 10-12 apps running simultaneously, and sticky notes scattered everywhere. Managers were trying to coach their teams by randomly scoring 2 out of 1,000 calls each month. It was clear that being an agent was an incredibly challenging job, and the high attrition rate of over 40% reflected that. This, in turn, led to poor customer satisfaction, long hold times, and agents transferring calls from one to another. Back in 2017, AI wasn’t yet “cool,” but we recognized its potential and the transformative impact it could have on Contact Centers. We understood that AI alone wouldn’t be enough; it needed to be paired with the right workflows and integrations. That realization became our secret sauce as we developed products to solve real problems for contact centers. If you had told me in 2017 that we would one day serve some of the world’s largest consumer brands, helping hundreds of thousands of agents and thousands of QA managers, I would have been skeptical. Yet here we are, with not just the best AI in the market, but an AI that’s deployed across 350+ customers, delivering truly impactful ROI: - $8 million in incremental revenue for a leading healthcare company - 50% reduction in after-call work at a leading healthcare technology company - 19% increase in quality scores and 8% reduction in AHT at a large HR people management software company - 23% reduction in AHT and 60% reduction in hold time violations at a leading financial services company - $2 million in annual operational savings at a leading personal finance technology company I am deeply grateful to my co-founders, customers, employees—both past and present, investors, and partners. Your belief and support have been the driving force behind turning our vision into reality. We could not have come this far without you. Thank you for being part of this incredible journey. Onwards and Upwards 🚀 https://lnkd.in/dZbuUNht

395

29 Comments

As models grow larger with the rise of LLMs, choosing the right precision type has become important. Sharing some insights over FP16 vs BF16:   🔷  FP16 and BF16 both consume same memory.  🔷  FP16 has better precision but lower dynamic range. In fp16, largest number is 65,504. BF16 can go upto 3.39e+38 !  This is because fp16 has higher # mantissa bits and bf16 has higher #exponent bits. 🔷  For model training, you will have to experiment and understand whether your network is sensitive to range or precision. If a model requires more dynamic range, it will converge better with bf16, which has same range as fp32. However, some models will converge better with higher precision (fp16) 🔷  When finetuning an fp16 model pre-trained in bf16, overflow issues may arise as some of the bf16 model weights might be larger than 65k.   Image source: nvidia   #precisiontype #fp16 #bf16 #modeltraining #LLMs #machinelearning #ai #datascience #torch

30

MLOps networking question: is there anyone based in India whose known for being great at MLOps? Like maybe they’ve - given some really great talks, or - done ML end to end (a few different ways/times), or - seen the full picture done at both small and LARGE data science teams, or - thought deeply about the various tools and where they fall short, and - are friendly and active in some of the communities - gone past the basics: anyone can train a model and get a batch inference pipeline or endpoint out the door—but few have thought about the hard problems like backfilling, monitoring, collaboration/exp tracking, governance, reproducibility, feature management, GPUs, working with lakehouses, data versioning, medium and large datasets, or lineage/data contracts I’m guessing they’d be a staff or principal engineer or maybe a lead / director. (Or undertitled) I can name a few people I know like this *outside* of India, but my network isn’t as strong there. (I’m thinking of Maria Vechtomova, Başak Tuğçe Eskili, 👨🏻‍💻 Jaco van Gelder) I’m hiring! It’s either a staff, principle, and/or tech lead. There's a team in Pune they'd be guiding/mentoring. I’m new to the team and the time difference in India makes communication harder so I want to find a friend I can trust and rely on to learn with me and make our ML Platform one of the best in the world. And also have fun with. We’re on Snowflake and AWS. Looking to hire a lot and select tools this year for the largest 3rd party seller on Amazon. We get to rebuild the platform from scratch for a really willing team in a profitable, private company of ~1600 where DS/AI/ML/BI is the star of the show right now. Growing the DS team from 20 to 30 this year. Powering several UI’s and major purchasing decisions. Tons of executive enthusiasm and support will be pointed our way. Currently 95% batch serving. Possibility of real-time on the horizon. Data: tabular, text, and images. The salary budget is aimed at India.

68

24 Comments

🌟 How to get started with GenerativeAi (RoadMap for Beginners) 🌟 💡Here’s a 5 Week RoadMap for anyone who would like to start their Generative AI Journey 1️⃣ Week 1: LLM Basics and Foundation Introduction to Large Language Models, covering their architecture, training techniques, and real-world applications. 📕Course: - Transformer course by HuggingFace - Introduction to GenerativeAi by DataCamp 2️⃣ Week 1-2: Prompt Engineering Exploration of Large Language Models, covering core concepts and advanced techniques for effective prompt design and optimization 📕Course: - ChatGPT PromptEngineering for Developers by DeepLearning.ai - Prompt Engineering with Llama2 by DeepLearning.ai 3️⃣ Week 2-3: Building LLM Application Developing powerful AI applications using Large Language Models and the versatile LangChain framework. 📕Course: - LangChain chat with your data by DeepLearning.ai - LangChain for LLM Application Development by Deeplearning.ai - Building Applications with Vector Databases by DeepLearningAI 4️⃣ Week 3-4: RAG and FineTuning Master the art of enhancing LLM performance through Retrieval-Augmented Generation and custom model fine-tuning for specialized tasks. 📕Course: - Augment your LLM using RAG by Nvidia - Knowledge Graphs for RAG by DeepLearning.ai - FineTuning LLM by DeepLearning.ai 5️⃣ Week 4-5: Evaluation & Implementation Putting your LLM model into production 📕Course: - LLM Ops by Deeplearning.ai - Building and Evaluating AdvancedRAG Applications by DeepLearning.ai (Link to courses in 1st comment) 🤔 The best way to update is to create real-life applications So here’s a Github repository containing GenerativeAI project :- (Link in 2nd comment) #GenAI #GenerativeAi #llm #chatgpt #learngenerativeai #generativeairoadmap #generativeaiguide #learngenai #ChatGPT  #chatbots #chatgpt4 #chatgpt3 #GenerativeAI  #ArtificialIntelligence #AI #MachineLearning #ml #AIroadmap #DataScience

31

4 Comments

I have been searching for the perfect tool that will do research and write a 3-5 pager report on any topic. Till now I have been very impressed by GPT Researcher https://lnkd.in/gBwJkjEp But now there is a new contender https://lnkd.in/gJHFCNrD Both these tools use a chain of Agents. Each task spawns multiple agents and once each agent is done, the tool summarizes the report from all the agents. (An Agent is a LLM doing a specialized task.). The Chain of Agents shows great promise beyond reports and can be used in multiple industries like Insurance, Finance, Stock trading etc. Tell me what you think ?

10

🚀 Exciting breakthrough in Indian AI! Introducing Sarvam-1, India's first open-source LLM built by Sarvam specifically for Indian languages! 🇮🇳 🔍 Key highlights: 🔹2B parameter model trained on 4T tokens high-quality data with 2T tokens of Indic language data  🔹Leveraged Llama 3.1 405B model to generate high-quality synthetic training data of ~2 trillion tokens across academic, technical & general domains, with content evenly distributed across the 10 supported Indian languages, except for Hindi, which constitutes about 20% of the dataset 🔹Superior token parity score (1.4-2.1 tokens per word vs 4-8 in other models) 🔹Languages supported: Bengali, Gujarati, Hindi, Kannada, Malayalam, Marathi, Oriya, Punjabi, Tamil, and Telugu + English 🔹Outperforms larger models like Gemma-2B and Llama-3.2-3B on key benchmarks 🔹4-6x faster inference compared to larger models 📈 Benchmark Performance: 🔹Outperforms Gemma-2B and Llama-3.2-3B 🔹Exceptional TriviaQA scores: 86.11% across Indic languages 🔹Strong MMLU (38.22%) and ARC-Challenge (46.71%) and IndicGenBench results 🔹Outstanding translation metrics: 46.81 chrF++ score on Flores 🔹Strong performance in both knowledge and reasoning tasks 🌟 Training Data : Document quality 🔹2x longer documents vs standard web data 🔹3x more high-quality samples Content distribution 🔹8x more scientific/technical content 🔹6x more programming documentation 🔹Balanced domain representation 💡 Technical Innovations: 🔹Ground-up optimization for Indian languages 🔹NeMo Curator domain and quality classifier models to improve data quality 🔹Custom tokenizer with revolutionary efficiency (1.4-2.1 tokens/word vs industry standard 4-8) 🔹Optimized architecture: 2048 hidden size, 16 attention heads, 8,192 maximum position embeddings 🔹Trained using NVIDIA's NeMo framework 🔹Trained on Yotta Data Services Private Limited's Shakti cluster using 1,024 GPUs for 5 days 💭 As someone passionate to build SLMs for India, I'm particularly excited about this development. What aspects of Sarvam-1 intrigue you the most? 🤝 Open Source Community: The model is open-sourced, inviting developers, researchers, and innovators to build upon this foundation. What applications do you envision building with Sarvam-1? Share your ideas below! 👇 Vivek Raghavan Pratyush Kumar Sunil Gupta #SLM #LLM #GenAI

41

1 Comment

Long-context LLMs are outperforming RAGs, so will RAGs be dead? 🤔 In this post, you will learn: 1. Comparison of RAG and LC LLMs 2. Performance - Cost Trade-off 3. Self Route: A new approach to bridge the gap What are your views on this research? Let me know in the comments! --------------------------------------------------------------------------- I hope this post helped you! Help people in your network by liking and reposting it ❤️ It inspires me to do more for our community. Follow Bhavishya Pandit to stay updated on GenAI! 🔥 #ml #llm #ai #genai #datascience #artificialintelligence #opensource #ibm #generativeai #largelanguagemodels #huggingface #openai #google #rag #machinelearning #python #langchain #microsoft #genai

516

49 Comments

Very common rookie mistake candidates do while trying to get a job in the data science industry. Having the following 3 projects from #Kaggle in your resume: House Price Prediction Telco Churn Prediction Titanic Problem No, you will not get your resume selected with these projects. I get it. YouTube tutorials suggested doing these projects so you are wondering, “Why aren’t my resume getting picked after I’ve completed them?” Well think about it. As a hiring manager, how do they know you didn’t just copy an existing notebook? You might say, “But they can at least give me a chance to prove my skills!” Again think from their perspectives - What if they do give a chance but you don’t perform up to the mark, that’s valuable time wasted. Now imagine giving that time to all candidates. To really get noticed, work on projects that will show that you’re not just another candidate. And there are ways. Many ways. We will discuss some during https://lnkd.in/gMNckgRi on 14th October, 9pm. #DataScience #DataAnalytics #HiringTips #CareerAdvice #DataJobs #Kaggle #ResumeTips #StandOut #DataProjects #JobSearch #Job

9

🤔 Interesting perspective from Infosys Chairman Nandan Nilekani: "Let Silicon Valley build LLMs; India should lead in AI use cases." At Meta's Build with AI summit, Nilekani suggested that India should focus on practical AI applications rather than competing in the LLM space. But is this the right approach? Key points to consider: • Silicon Valley giants are heavily invested in LLM development • India could focus on: - Synthetic data generation - Small language models - Real-world applications - Building robust data infrastructure The vision is for India to become the global "use case capital" for AI, emphasizing frugal innovation and practical solutions. What do you think? 💭 Should India stick to developing practical AI applications? 💭 Or should we also invest in fundamental AI research and LLM development? 💭 Could focusing solely on use cases limit our AI capabilities in the long run? Let's discuss! Share your thoughts in the comments below. 👇 #AI #DigitalIndia #LLM

30

DISRUPT-DS industry roundtable IV - Large Language Models 🔨 Following up on my previous posts on the DISRUPT-DS roundatable, there is a benchmark on the potential of LLMs on R&D. Unlike the overall potential of Data Science which is considered disruptive, surprisingly the conclusion around LLMs is that they will only have a moderate impact via efficiency gains. And even more more suprisingly, the most impacted R&D area is ... 🔫 ... Data Science itself and purportedly areas where one writes computer code. However, given the fact that most LLMs can write and execute decent code only in Python, it leaves many statistical sciences in the lurch. 🔍 The next major area of impact is insight generation via searching and summarising. Whether it be searching internal regulatory documents via custom-built RAG pipelines or being able to query *both* data and documents systems simultaneously through agentic-applications. The ability to browse multiple systems, both internal and external to generate insightful and (hopefully) factual summaries remains the most sought-after use-case. ✍ Document-writing by summarizing larger reports or data from disparate sources, which might seem the most intuitive use-case for Generative AI, fails to get sufficient traction in R&D and lands only on the third spot. The reasons are the constraints imposed by regulatory writing, fact heavy document-authoring requiring complex agentic-designs which are often unstable and a work-in-progress. 🏭 Automating some of the R&D workflows by combining the above mentioned agents which can write and execute code, search multiple systems and summarise information in factual content is the most complex use-case. Consequently, only a quarter of Pharma companies from the benchmark consider it as realistic prospect. LLMs remain one of the most exciting area in AI and the potential is immense. However, in Pharma R&D the current state-of-implementation is still in its infancy. I plan to dedicate a series of posts in the future on this topic about my experiences of building solutions in this space.

17

2 Comments

Quantized embeddings offer significant advantages for NLP models, including reduced storage requirements, faster computations, lower energy consumption, and improved model interpretability. They are becoming increasingly important for enabling efficient and scalable NLP on resource-constrained devices and cloud-based applications. Read my latest article below

6

My latest Medium post comparing SetFit, FastFit, and Semantic Router for Few-shot NLP Intent Classification has been published by Towards AI! This post covers: 1. Details of NLP Intent classification and why it is extremely important in the age of LLM chatbots 2. Algorithm details of SetFit, FastFit and Semantic Router 3. Detailed code experiments on a chatbot intent detection dataset 4. Conclusion on the best method as of today. As with most of my posts, it has lots of code you can directly use in your work! This post was jointly written by Shilpa and me, both of us humans 😅 🔗 Medium Friend Link (without paywall): https://lnkd.in/dVfAvjzr Follow me for more tips on building successful ML and LLM products! Medium: https://lnkd.in/g2jAJn5 X: https://lnkd.in/g_JbKEkM #generativeai #llm #nlp #artificialintelligence #mlops #llmops

78

9 Comments

Hugging Face updated its OpenLLM leaderboard to use harder benchmarks and there are some really interesting findings: 🚫 Models have gotten so good at high-school-level benchmarks they're no longer useful. Some model developers may also have explicitly trained to do well on these benchmarks, hence now do poorly on harder ones. 🔐 Closed models are way better on tough problems. The best GPQA (graduate-level questions) performance of any open model is 17.23. Claude 3.5 Sonnet gets 59.4 0-shot on the toughest set of questions. 🎓 Small models can do very well. Microsoft's tiny Phi-3 medium model—a 3.8B parameter model—performs nearly as well as models over 18X as large, and beats Lllama-3 70B on benchmarks including GPQA. This is quite surprising and suggests that the approach to training Phi models—using high-quality synthetic data generated by larger models for a learning curriculum—is very fruitful. 🇨🇳 Chinese models are doing very well. The idea that China is far behind the US in AI development no longer seems to be true. Leaderboard is here: https://lnkd.in/gv92Z5CR

3

Hello Network ! Hello Tunisia Tech and Specially AI community ! I am honored to be speaking at Tunisia Dev Days 2024 🇹🇳 (PREMIER .NET & AI CONFERENCE IN TUNISIA 🇹🇳 ) , happening from November 7th to 9th at Mövenpick Hotel du Lac Tunis . As an #AI and #XAI expert, I’ll be joining this incredible lineup of world-class tech innovators and leaders for three days packed with groundbreaking insights, hands-on sessions, and deep dives into the most cutting-edge technologies in #AI, #NET #cloud_computing, #software_development, and more. My session: "XAI Unleashed: Bringing Clarity to the Complex World of AI" will dive deep into the importance of transparency and interpretability in AI systems. As AI becomes more integral to business decisions, auditing, healthcare, and more, the need to understand why AI makes certain decisions is crucial and I'll explore real-world use cases ( stay Tuned for them !!). It’s going to be a fascinating dive into how AI models can become more transparent and trustworthy, shaping the future of AI for the better. But I’m not The only Expert ! Tunisia Dev Days is set to feature a powerhouse lineup of speakers leaders in their fields from the Biggest companies Around the World 🌎 who will bring unique insights and cutting-edge knowledge across a wide array of topics. These amazing thought leaders will be joining me to explore #AI, #Machine_Learning, #Cloud_Computing, #DevOps, #Quantum_Computing, and much more over the course of this three day event. It’s an exciting opportunity to network with industry experts, dive into the latest tech trends, and discuss how we can continue driving innovation in Tunisia and beyond. The most important Question is Why attend? Easily it’s for :Gaining practical insights from AI experts ,Exploring new trends in #Cloud_Development, #AI Ethics, #Quantum_Computing, #Agile_Methodologies, and more ,Attending hands-on workshops to boost your technical skills and engage in thought-provoking discussions  and Networking with professionals across diverse sectors, from tech startups to global companies Whether you're an AI enthusiast, developer, business leader, or just curious about the future of technology, this event will have something for you! I can’t wait to be a part of this transformative event and share my thoughts on the role of #Explainable_AI (XAI)in making AI systems more transparent and understandable. 📅 Mark your calendars!  📍 Tunisia Dev Days 2024 | November 7th-9th   🌐Learn more and register here at Tunisia Dev Days Website : https://lnkd.in/gaRmuc4i See you there everyone ! Let’s connect, exchange ideas, and shape the future of AI together! See you there! #TunisiaDevDays #XAI #AI #TechInnovation #AITransparency #ExplainableAI #Networking #TechLeadership #NET #AIForGood #AITrust

61

Addresses fundamental challenge of integrating contextual domain knowledge with RAG. Essential for real life applications in AI.

15