Gaurav Agarwal’s Post

My second medium article! How to predict employee/customer attrition? ☠ Building machine learning pipelines can be daunting for beginners, so I wanted to document some interesting pointers I learned along the way. This article explores different approaches to constructing ML pipelines, based on a project I completed two years ago. A big thank you to AI Singapore for providing the dataset—I hope it's okay to use it! Writing this and revisiting my past projects has reminded me of the fascinating potential of data. It can be used to identify new business opportunities, predict actions to prevent scams or accidents, and much more. Do hit me up if you need someone to explore data with! always looking for like-minded friends :D 👨🔬 😍 #MachineLearning #DataScience #MLPipelines #AI #DataAnalytics #TechJourney #Innovation #DataProjects #DataExploration

Sharing this story with my network as it explains potential LLM use-cases, and syntactic data, in such a clear way. Enjoy Christmas Eve is an operational nightmare. This year, a team of Elves used Foundation Model APIs + Llama2 with #MosaicML Inference to create an AI model for analyzing children’s letters to Santa and identifying the right present! Find out how they did it by leveraging techniques like synthetic data generation and name entity recognition with the Databricks Data Intelligence Platform. #syntacticdata

Looker Dashboard with Actionable Threat Intelligence, from FULLY ENCRYPTED data. Visit OmniIndex at stand B156 Infosecurity Europe to get a live demo of our fully homomorphic encryption making AI insights from fully encrypted data possible. In real-time! #AI #Datasecurity #encryption #looker #logfiles #fullyhomomorphicencryption

If a man with two clocks doesn't know what time it is, #GenAI with two clocks doesn't know what time it is at scale. Fix your #data. Only then (to quote Tom Godden) will you be able to do anything more than clever parlor tricks with #GenerativeAI.

Quite remarkable on what you can do with Local LLMs these days hosted on your own desktop. I fed Llama3 7B model a list of Finnish companies by market cap in CSV format and prompted it to generate a pie chart. It provided the results without any additional code execution on the user side. Chatting with your data is becoming a thing!! #llm #llama3 #chatwithyourdata #datanalytics #pypandas #ai

Looking forward to talking about how Alation is using AI to help you find the right information faster

👇#Post16 𝐨𝐟 𝐦𝐚𝐜𝐡𝐢𝐧𝐞 𝐥𝐞𝐚𝐫𝐧𝐢𝐧𝐠 👉𝐀𝐝𝐯𝐚𝐧𝐭𝐚𝐠𝐞𝐬 𝐨𝐟 𝐋𝐨𝐠𝐢𝐬𝐭𝐢𝐜 𝐫𝐞𝐠𝐫𝐞𝐬𝐬𝐢𝐨𝐧 𝐦𝐨𝐝𝐞𝐥: 1)Simple and interpretable 2)Probabilistic predictions 3)Works well with linearly separable data 👉𝐃𝐢𝐬𝐚𝐝𝐯𝐚𝐧𝐭𝐚𝐠𝐞𝐬 𝐨𝐟 𝐋𝐨𝐠𝐢𝐬𝐭𝐢𝐜 𝐫𝐞𝐠𝐫𝐞𝐬𝐬𝐢𝐨𝐧 𝐦𝐨𝐝𝐞𝐥:: 1)Limited to linear decision boundaries 2)Sensitive to outliers 3)Assumes independence of features Feel free to drop any advice or resources in the comment below #learnings #datascience #machinelearning #logisticregression

People are always looking for that shining new "thing" that will solve their problems. I guess the logic being that if something old was good enough we would have seen the results. But there is another option - maybe we just don't do what we're supposed to do? Take something very simple for example - logging. We all know that we should do it, but we don't, and then complain that it's hard to troubleshoot, and then shop for AI based monitoring tools that will magically compensate for lack of logs. Right... #software_development #database_development #ifpthenq

ChartGen AI: Your data, your chart, your way! 📊 Simply upload your dataset, describe the chart you want, and let ChartGen AI do the rest. Try a random example or specify your dataset format - CSV, Excel, JSON, or link to Google Sheet. Streamline your data visualization effortlessly! #FREE #ChartGenAI #DataVisualization #AIAnalytics Larissa Abi Nakhle

#AI #ML #Tech Build your own RAG and run it locally on your laptop: ColBERT + DSPy + Streamlit: Tutorial for GenAI beginners: let’s build a very simple RAG (Retrieval Augmented Generation) system locally, step-by-step. Continue reading on Towards Data Science » #MachineLearning #ArtificialIntelligence #DataScience