Agilytic

🌍 We are happy to report our 1% For The Planet certification for the third consecutive year. 🌱 In this interview, Julien Theys recaps the reasons behind our commitment, and provides suggestions if your organization is considering joining the movement. (Spoiler alert: we highly recommend it 😇) #OnePercentForThePlanet #BeOnePercentBetter #certification

🎓 Agilytic est très heureux de contribuer à former les experts data de demain grâce à ce programme du SMCS (Statistical Methodology and Computing Service) - UCLouvain

💡 Many data specialists find themselves rethinking their roles in the data world. John Hammerschlag is one of them. 🧐 Drawn initially to data science for its thrill of discovery, John has since shifted his focus to data engineering—a path that, as he puts it, allows him to ‘build the road before the journey begins.’  📰 Learn more about his journey and findings in this interview: https://lnkd.in/eS7A28nU

Client: “We do have clean data for this project.” Me: “How much do you bet on that?” My colleagues at Agilytic will all tell you the same: in the world of data analytics, “clean” data often turns out to be just surface-level clean. Behind the scenes, a surprising amount of data fails essential quality checks, which can lead to unexpected challenges in advanced analytics and AI projects. 🗑️ In fact, one of the biggest bottlenecks in analytics isn’t the modeling—it’s the data itself. 𝗚𝗮𝗿𝗯𝗮𝗴𝗲 𝗶𝗻, 𝗴𝗮𝗿𝗯𝗮𝗴𝗲 𝗼𝘂𝘁 is a core truth that only grows in importance as analytics advance. 👌 Data quality is multi-dimensional, with several aspects that need to be evaluated carefully before diving into any modeling. We very often encounter the following three: -- 𝗖𝗼𝗺𝗽𝗹𝗲𝘁𝗲𝗻𝗲𝘀𝘀 – Are all necessary data points present? Missing sales channels, for instance, can biaise results and limit insights. -- 𝗖𝗼𝗻𝘀𝗶𝘀𝘁𝗲𝗻𝗰𝘆 – Does the data align across sources and time periods? Inconsistent data leads to conflicting findings. This is often the case with dates that are not following the same definition over time. -- 𝗔𝗰𝗰𝘂𝗿𝗮𝗰𝘆 – Is the data correct and reliable? Even a small percentage of inaccuracies can throw off predictions. This typically happens with data that is encoded manually. Each of these factors can drastically impact the outcomes of analytics projects, making it crucial to vet data against these dimensions before moving forward. In our era of 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗔𝗜 and 𝗥𝗔𝗚 (Retrieval-Augmented Generation) tools, ensuring quality data has become even more essential. Poor data quality can lead not only to errors but to hallucinations—where AI systems confidently deliver inaccurate or misleading results. Ultimately, high-quality data isn’t just a nice-to-have; it’s essential for reliable, actionable insights. By investing time in comprehensive data checks, organizations can significantly improve the accuracy and value of their analytics outputs, ensuring AI models reflect reality rather than illusion. What’s your experience with data quality in analytics? Have you encountered unexpected challenges or surprises? Let’s hear your thoughts below! #dataquality #analytics #predictions #generativeAI #RAG

What’s Next for AI in Healthcare? Here’s What I Learned from Clinicians. As someone who’s been deep in the world of data and AI for years, I’ve seen technology transform industries. But healthcare? That’s a different story, at least to me. Recently, I attended an event from WeLink.Care on the theme “L’IA en santé au-delà du mythe”, and it was one of the most inspiring—and eye-opening—experiences I’ve had in a while. The conversation went beyond tech, diving into what really matters for AI to make a meaningful impact on patient care. In the article below I share the key insights that stuck with me, that came directly from a panel session with clinicians. Thank you Emilie Fockedey and Digital Wallonia for the invitation. Enjoy the reading. #AIinHealthcare #HealthTech #DataDrivenHealth #DigitalTransformation #FutureOfHealthcare

💡 “We need AI because... well, everyone’s doing it!” - Ever heard that phrase ? Yep, that’s exactly what a client’s ExCo said to me the other day. And they are not the only ones, I met a few of them in the last months. I get it—AI is all the rage. But here’s the thing: AI isn’t a magic wand you wave around just because it’s trending. As a consultant in Data & Analytics, my response is essentially something like, “Cool, but what’s the problem we’re actually trying to solve?” Sure, LLMs applied for Intelligent Document Processing or Knowledge Base Search can be game-changers for the right use case. But for most companies, a solid report based on good qualitative data or a tried-and-true classification algorithm works wonders. Sometimes, the simple solution is the best one. The point is: don’t jump on the AI bandwagon just because everyone else is. Focus on solving the real problems. AI should be a means to an end, not the end itself. #AI #AnalyticsConsulting #GenAI #BusinessStrategy #ProblemSolving #KeepItSimple

🥳 We have a new colleague! 🏁 Wishing Loïc Meunier a warm welcome to Agilytic! 🎓 Armed with an MSc in Computer Science with a major in Computer Systems Security, Loïc's journey began with an internship at Deloitte in their SAP Data & Analytics team. There, he developed an internal screening tool using LLMs. ⛹️♂️ Standing at an impressive 2.10 meters, Loïc is literally impossible to miss. When he is not towering over the office, you are likely to find him on a basketball or padel court. 👨🏫 He is also known for diving deep into fascinating (read: obscure) subjects.

🚀✨ When the brilliant Stephen Wolfram publishes a didactic, illustrated essay on machine learning and neural networks, you know it’s time to dive deep! 📚🤖

🎓 Are you freshly graduated, and aspiring Data Engineer or Data Scientist? 🚀 Eager to keep learning, while working on challenging, impactful projects? ✨ Don't miss our onboarding month this October! #werehiring #dataengineering #datascience

🔢 Nettoyer les données, les analyser, extraire l’information et la communiquer: voilà des compétences indispensables, et donc très recherchées dans le monde du travail. 🎓 C'est là précisément ce que le SMCS (Statistical Methodology and Computing Service) - UCLouvain vous propose d’apprendre dans le cadre de la formation en "Junior #Data Analyst" ! 💪 Une formation qui peut compter sur le soutien d'Agilytic