JMP

One challenge I kept hearing from customers at a recent JMP event was that even with plenty of data we often struggle to clearly understand how complex systems behave when variables start interacting with each other. That’s where modeling and visualization can make a real difference. Instead of only reviewing outputs after the fact, predictive modeling helps teams explore relationships, test “what if” scenarios, and make decisions with greater confidence before moving forward. I recently watched this on-demand session demonstrating how modeling tools – including the Prediction Profiler – can help make complex, noisy, or time-dependent processes easier to interpret and communicate. Worth a watch if you’re looking to move from raw data toward faster, clearer decision-making. Curious how others are approaching this: Are you leaning more on predictive models to guide decisions, or still evaluating outputs case by case? #PredictiveAnalytics #MachineLearning https://lnkd.in/gqHzcphJ

I’ll be speaking at SEMI Americas Advanced Semiconductor Manufacturing Conference this week. My session, “Hybrid Experimentation: Improving the Efficiency of Bayesian Optimization with Definitive Screening Designs,” will focus on practical ways to: Get more insight from fewer experiments Combine structured DOE with adaptive optimization Reduce guesswork in complex process development 📍Hilton Albany - Governor AB 🕒 May 13 at 2:40 pm If you’re attending, I’d enjoy connecting. Feel free to stop by or reach out.

So proud of Kaylee! What an accomplishment!!!

Looking forward to SEMI Americas Advanced Semiconductor Manufacturing Conference this year. It’s always a great opportunity to step away from day-to-day work and connect with others tackling similar challenges in semiconductor process development and optimization. I’ll be presenting “Hybrid Experimentation: Improving the Efficiency of Bayesian Optimization with Definitive Screening Designs,” where I’ll share some practical approaches to getting more insight from fewer experiments. If you’re planning to be there, I’d really enjoy catching up - whether during the session or in between talks. Hope to see some familiar faces (and meet a few new ones as well). https://lnkd.in/g43UVqiK

QUIZ TIME: Which of these Screen Sizes should the team stop using? Let me know in the comments. I made this graph in 20 seconds and only used drag and drop functions of JMP to align the data. Unless an AI starts reading my mind, I could not have been faster. This is why we are putting on a 2 day workshop on the 28th and 29th of April to show you how to do visualisation with JMP. Registration is free and there is genuinely a lot of great stuff to learn! The event will be held by Ben Barroso-Ingham and Spyros Megalou , who are two of our very own data wizards. Sign up here or pm me for more information: https://lnkd.in/d_rRC-WJ If you need JMP for the workshop, I am also happy to give you a free trial! #JMP #datavisualisation #analytics #statistics #biotech #manufacturing

Today it's #InternationalBookDay , but for those of us from Catalonia, it’s also one of the most meaningful cultural traditions of the year: Sant Jordi. A day filled with books, roses, and the joy of sharing stories. Just like the tradition of exchanging books and roses, our JMP Book Club brings people together through shared learning and curiosity. 📘🌹 Want to take a look or join the conversation? Here’s the link: https://lnkd.in/dHfDpZUS Feliç Sant Jordi! Anne Milley , Di Michelson , Maria Lanzerath , Gail Massari , Luiza Fabreti, PhD , Natalie Starkey

I'm working on lots of new mandates, so if you are currently open to new opportunities...lets chat! And if you don't want to chat, then scroll through my postings here and hit me up if you're interested in hearing more about any of them. Please keep in mind, these are all Montreal based opportunities (except for one NY based Sales position), so if you live outside of Canada please scroll on.

Working with scientists and engineers across industries for several years, I see very clearly what benefits Smart Experimentation methods promise and deliver. Software and algorithm are rarely the problem. The challenge rather comes from real world characteristics, which prohibit simple one-tool-solves-it-all solutions. More so, it’s down to the right toolset, and culture. From my experience, four domains need to be navigated in order to effectively deploy smart experimentation methods. I call them the 4 P's - and I presented this framework at the JMP Discovery Summit Europe last month. Here’s a brief summary: Purpose. What is the actual goal? BayesOpt is purpose-built to reach a defined optimum fast. If the goal is "understand the process" rather than "find the best conditions," it's the wrong primary tool. A thorough consideration of “What do we want” is important. Prior. What data already exists? DOE is the perfect match if starting from nothing. BayesOpt with no prior data will spend its first few runs collecting information before it starts optimising. BayesOpt seeded with a well-designed DOE - or even messy historical data - converges far faster. The quality of what you put in directly determines the speed you get out. Problems. What are the constraints? Factor constraints, execution limits, process variability, shifting project scope - these don't disappear when you change the algorithm. But BayesOpt handles them differently from DOE (half a DOE is pretty worthless). And BayesOpt is more agnostic to the true shape of the relationship, and it navigates constrained spaces that classical designs may struggle with. People. Who is running this, and what do they expect? We cannot change how people feel about things like statistically designed experiments or gaussian process models, but we can use different tools for similar tasks helping people overcome their hurdles and start using smart experimentation methods. People are THE central part if companies try to achieve a cultural shift and transform industrial experimentation. I am humbled by the feedback I got after my talk – it confirmed how similar the challenges are companies are facing these days. I’d love to continue the conversations - Which of these four do you think causes the most friction in your context? #bayesianoptimization #smartexperimentation #R &D #chemicalindustry #experimentaldesign #processoptimization #innovation #leadership #JMP (Again, thanks to Florence Kussener :))

Lay-offs, missed quota, becoming irrelevant in a crowded, AI-hyped market. Common fears among sales people right now. Not so at JMP. Join a winning team and apply for this Account Executive role if you are UK or Denmark based: https://lnkd.in/dmnSauh6 Danielle Rogers Hadley Myers Willem Clow Emmanuel ROMEU #Sales #Job #JobOffer #Hiring