iiinova — where your digital self keeps evolving

Chris Rackauckas @chrisrackauckas.bsky.social · 21/07/2026

GPT 5.6 or Claude Fable, which is better for Physical AI? We put it to test in the #Dyad AI evaluation system for generating models and tuning controllers. Our evals gave pretty definitive results, check it out! juliahub.com/blog/frontie... #AgenticAI #AI4Science #SciML #Julia

juliahub.com

GPT-5.6 vs Claude Fable 5 for Physical AI, which performs best? - Blog | JuliaHub

We tested the latest frontier models in the Dyad agent on five modeling and simulation problems, comparing accuracy, cost, time, and work style.

031

Chris Rackauckas @chrisrackauckas.bsky.social · 12/04/2026

Real-world telemetry data and Internet of Things collide with scientific machine learning #sciml? Devices which can learn how to adapt to reality and continually improve their efficiency. #UDEs with #Dyad => better thermal efficiency. See : arxiv.org/abs/2603.15943 #julialang #AI4Science

arxiv.org

Scientific Machine Learning-assisted Model Discovery from Telemetry Data

Calibration of dynamic models to data is an important step in building building digital twins of HVAC equipment, thermal loads and control systems. Sometimes, when a model fails to calibrate to data, ...

070

Chris Rackauckas @chrisrackauckas.bsky.social · 17/03/2026

Peer reviewed result: better languages leads to better agentic AI and LLMs. A language where if you connect things that have incompatible units you get compiler errors for the agents to quickly correct. This language of AI for Science is #Dyad. www.sciencedirect.com/science/arti...

sciencedirect.com

Improving process systems engineering with specialized multi-agent large language models

Multi-agent large language models (LLMs) were evaluated for Process Systems Engineering (PSE) tasks. Two multi-agent systems (Dyad 2.1 and Claude Opus…

341

Chris Rackauckas @chrisrackauckas.bsky.social · 27/02/2026

Join us for a Dyad Modeling Livestream today - this time at 1pm ET / 10 am PT! Michael Tiller will joining us today to model a hybrid-EV powertrain! Tune in on YouTube and send us your thoughts in the chat! www.youtube.com/watch?v=qLfV... #sciml #julialang #dyad

youtube.com

Dyad AI Modeling Challenge: Modeling a Hybrid EV Powertrain

YouTube video by JuliaHub

341

Chris Rackauckas @chrisrackauckas.bsky.social · 26/02/2026

Recent optimizations in SciMLSensitivity.jl are having some pretty good payoffs! Given we just 2.5x'd 2 months ago, this next change of 3.2x is putting us almost an order of magnitude ahead! See the latest autodiff benchmarks docs.sciml.ai/SciMLBenchma... #sciml #julialang

291

Chris Rackauckas @chrisrackauckas.bsky.social · 05/02/2026

New fastest explicit non-stiff ODE solver? That's right, we now have something beating the pants off of the high order explicit RK methods! Check out the new symbolic-numeric optimized Taylor methods available in DifferentialEquations.jl! #julialang #diffeq #sciml

181

Reposted by Chris Rackauckas

JuliaCon 2026 @juliacon.org · 28/01/2026

@juliacon.org 2026 will have a minisymposium on "Methods and Applications of Scientific Machine Learning (SciML)" hosted by @chrisrackauckas.bsky.social find out more on pretalx.com/juliacon-202... and submit your proposal through juliacon.org/2026/cfp/ #julialang #sciml

Scientific machine learning (SciML) methods are techniques which incorporate machine learning with mechanistic modeling. The purpose of this minisymposium is to share improved methods and applications of SciML to showcase the ever advancing ecosystem in Julia.

Examples of topics fit for this minisymposium include:

New algorithms and software for physics-informed neural networks
Tips, tricks, and techniques for improving convergence of universal differential equations
Applications of fitting universal differential equations on real data and verification
Methods which use neural networks with classical solvers in new ways
Advancements in automatic differentiation for SciML applications
Methods which use learning to accelerate numerical simulations (surrogates, metamodels, emulators, reduced order methods (ROMs))
Advancements in compiler and optimization techniques to improve learning in SciML scenarios.

084

Chris Rackauckas @chrisrackauckas.bsky.social · 14/01/2026

Your college professor teaches you "A-stable methods are required for stiff ODEs". But PSA, the most commonly used stiff ODE solvers (adaptive order BDF methods) are not A-stable. #sciml #numericalanalysis #diffeq www.youtube.com/shorts/hmKVQ...

youtube.com

The most commonly used stiff ODE solver isn't A-stable?!?!?

YouTube video by Chris Rackauckas

011

Chris Rackauckas @chrisrackauckas.bsky.social · 13/01/2026

Physics-Informed Neural Surrogates for Mesh-Invariant Modeling of High-Speed Flows at #AIAA #SciTech! We built a neural surrogate that predicts aerodynamic behavior 595x faster than CFD while maintaining ~1% relative error. #sciml #Julialang #CFD #Hypersonics #AIAASciTech

170

Chris Rackauckas @chrisrackauckas.bsky.social · 09/12/2025

The #Julialang Dispatch podcast wrapped said we were in the top 20% of videos on on Spotify this year! Also it was a heavily marathoned show, meaning "Fans listened to you for longer than 91% of shows" Looks like everyone is staying to the end 😘.

080

Chris Rackauckas @chrisrackauckas.bsky.social · 08/12/2025

Scientific machine learning (#SciML) is not just about adding scientific information to machine learning, but it's also about making machine learning accessible to heterogeneous data. youtube.com/shorts/jop2S...

youtube.com

Scientific machine learning and heterogeneous data

YouTube video by Chris Rackauckas

031

Chris Rackauckas @chrisrackauckas.bsky.social · 28/11/2025

New livestream, #Dyad Modeling Live! In this stream we built up a thermal model of a room using #AgenticAI and added a heat pump with different control strategies and analyzed the power efficiency. Join the fun live next week! #julialang #sciml youtube.com/live/I542x6g...

youtube.com

Dyad Modeling Live: Creating a Thermal Model of a Room using Agentic AI

YouTube video by JuliaHub

121

Chris Rackauckas @chrisrackauckas.bsky.social · 25/11/2025

How to properly cook your turkey, using agentic AI and #dyad #julialang! Happy Thanksgiving from JuliaHub, hope we can help you with acausal modeling to change your family's holiday! youtube.com/shorts/qv6Qv...

youtube.com

How to properly cook your turkey using agentic AI and #Dyad #julialang!

YouTube video by JuliaHub

041

Chris Rackauckas @chrisrackauckas.bsky.social · 25/11/2025

A lot of people think JIT compilers, or the compilation process in general, magically makes programming languages faster. "Just add a compiler", nope that's voodoo. So why are "traditional compiled languages" like C and Fortran faster? #JuliaLang youtube.com/shorts/Z1_vZ...

youtube.com

Do (JIT) compilers make code faster? Not always, it's all about program information! #julialang

YouTube video by Chris Rackauckas

081

Chris Rackauckas @chrisrackauckas.bsky.social · 19/11/2025

ANSYS /Synopsys, one of the largest simulation companies in the world, is partnering with @JuliaHub_Inc in order to bring #Dyad, #Julialang, and #SciML to next level of adoption. We have many things planned. This is how research becomes reality. www.prnewswire.com/news-release...

prnewswire.com

JuliaHub Partners with Synopsys to Power SciML-Based Digital Twins

/PRNewswire/ -- JuliaHub, a leader in AI-native simulation and modeling, today announced a strategic partnership with Synopsys (NASDAQ: SNPS) to integrate...

080

Chris Rackauckas @chrisrackauckas.bsky.social · 17/11/2025

#SciML fact of the day: automatic differentiation fails to give the correct derivative on a lot of very simple functions 😱 😱 😱 . #julialang #automaticdifferentiation youtube.com/shorts/KTguZ...

youtube.com

Automatic differentiation is incorrect on very simple functions??? 😱 😱 😱

YouTube video by Chris Rackauckas

1111

Chris Rackauckas @chrisrackauckas.bsky.social · 13/11/2025

Can Agentic AI turn single purpose code into reusable modular code? Dyad's specialized AI can! Watch our latest video on AI-assisted model restructuring and physics enhancement: www.youtube.com/watch?v=0RdA... #ModelingAndSimulation #AIAgent #JuliaLang #SciML #Dyad #SystemsEngineering #Modelica

youtube.com

Dyad Agentic Workflows: Generating Hierarchical Models and Increasing Physical Fidelity

YouTube video by JuliaHub

020

Chris Rackauckas @chrisrackauckas.bsky.social · 07/11/2025

Check out the latest #Julialang Dispatch Podcast! This episode goes over the changes to the Julia language, its standard libraries, and the main changes in the package ecosystem (#sciml, JuMP, etc.) that occurred over the summer. Most posts coming soon as well! www.youtube.com/watch?v=q3_W...

youtube.com

This Month in Julia World - August 2025 | JuliaCon Recap & Community Updates

YouTube video by Julia Dispatch

052

Chris Rackauckas @chrisrackauckas.bsky.social · 06/11/2025

Watch Dyad's AI agent build a complete thermal model from just an image! Picture -> validated DAEs in minutes. Features: Auto parameter generation, model optimization, custom animations. All with production-ready Julia code. youtu.be/eKLDVCkJC1s #dyad #julialang #sciml

youtu.be

Generating a Coffee Cup Thermal Model from a Schematic

YouTube video by JuliaHub

030

Chris Rackauckas @chrisrackauckas.bsky.social · 28/10/2025

#Dyad #SciML tutorial! Use Dyad's graphical/textual #acausal system to build models from validated model components and transform into your #digitaltwin! #Dyad = component-based modeling tool (e.g. #Modelica, #Amesim, #Simulink) + AI/ML autocomplete! www.youtube.com/watch?v=ttQI...

youtube.com

010

Chris Rackauckas @chrisrackauckas.bsky.social · 20/10/2025

What is #acausal modeling and how does it lead to better reproducibility and modularity in modeling and simulation? Check out this video which goes step-by-step into building acausal models using the RC circuit and RLC circuit www.youtube.com/watch?v=rMb4... #julialang #dyad #sciml

youtube.com

Acausal vs Causal Modeling Explained with Dyad

YouTube video by JuliaHub

021

Chris Rackauckas @chrisrackauckas.bsky.social · 16/10/2025

#LLMs make mistakes. Modeling languages like #Dyad have static analysis to compile-time check whether models are physically possible. What happens when you mix the two in an #agentic workflow? Automated construction of accurate models! www.youtube.com/watch?v=hIkb... #julialang #sciml

youtube.com

Agentic modeling of a friction brake with Dyad, LLMs for Model-Based Systems Engineering

YouTube video by JuliaHub

142

Chris Rackauckas @chrisrackauckas.bsky.social · 09/10/2025

SciML Developer Chat Episode 1: Base Splits and Symbolics Precompilation Welcome to the first episode of the SciML Dev Chat! We discuss the latest developments in the #SciML (Scientific Machine Learning) ecosystem for #julialang! youtu.be/0yQ4aZ-ABhY

youtu.be

SciML Developer Chat Episode 1: Trimming Support and Symbolics Precompilation

YouTube video by The Julia Programming Language

070

Chris Rackauckas @chrisrackauckas.bsky.social · 08/10/2025

MIT Julia Lab: looking for postdoctoral researchers in #julialang open source software development, scientific machine learning (#SciML), and systems biological / pharmacological modeling (#QSP) for research in equation discovery for large stiff systems. julia.mit.edu/projects/#po...

julia.mit.edu

Postdoc Projects

0113

Chris Rackauckas @chrisrackauckas.bsky.social · 03/10/2025

AI-Assisted Modeling in Dyad with Cursor AI! We've just released a comprehensive demonstration showing how LLMs can dramatically accelerate dynamic system modeling workflows in Dyad. youtu.be/awwJPWIlXXM?...

youtu.be

Demonstration of AI-Assisted Modeling in Dyad with Cursor AI

YouTube video by JuliaHub

030

Reposted by Chris Rackauckas

MACSYS @macsys.bsky.social · 30/09/2025

It's been a thrill to host Dr Chris Rackauckas for a 2-day visit to our node at QUT. Chris presented a seminar, ran a workshop on High-Performance Scientific Modelling with Julia & SciML, & met with our researchers. Chris is VP of Modelling and Simulation at JuliaHub. @chrisrackauckas.bsky.social

012

Chris Rackauckas @chrisrackauckas.bsky.social · 22/09/2025

Scientific Modeling Cheatsheet: #MATLAB vs #Python vs #Julialang Side-by-side comparison for: • Differential equations • Optimization • Automatic differentiation • Symbolic computing • More! Highlights ecosystem differences & best practices. sciml.github.io/Scientific_M...

sciml.github.io

Scientific Modeling Cheatsheet – MATLAB – Python – Julia Quick Reference

13514

Chris Rackauckas @chrisrackauckas.bsky.social · 22/09/2025

New workshop materials on High-Performance Scientific Modeling with Julia & SciML: • #Julialang for scientific computing • ODEs/PDEs & numerical methods • Symbolic-numeric modeling • Biological systems (Catalyst.jl) • Parameter estimation • #SciML & UDEs github.com/SciML/Julia_...

github.com

GitHub - SciML/Julia_Modeling_Workshop: High-Performance Scientific Modeling with Julia and SciML

High-Performance Scientific Modeling with Julia and SciML - SciML/Julia_Modeling_Workshop

083

Chris Rackauckas @chrisrackauckas.bsky.social · 19/09/2025

DifferentialEquations.jl is many things, and lots of people only use a small portion of it. Check out the JuliaCon 2025 workshop: introduces many aspects of the packages that the developers feel are underutilized and under-understood! #julialang #sciml www.youtube.com/watch?v=lSGF...

youtube.com

A Deep Dive Into DifferentialEquations.jl | JuliaCon Global 2025 | Rackauckas, Smith

YouTube video by The Julia Programming Language

0113

Reposted by Chris Rackauckas

MACSYS @macsys.bsky.social · 19/09/2025

Dr Chris Rackauckas will present a seminar at all three university stops: "Automating Construction of Biochemical and Physical Models with Scientific Machine Learning (SciML)" 📆Tues, 23 Sept: University of Melbourne 📆Thurs, 25 Sept: UNSW 📆Mon, 29 Sept: QUT 👉https://macsys.org/event/

011

Reposted by Chris Rackauckas

MACSYS @macsys.bsky.social · 19/09/2025

MACSYS is excited to host Dr Chris Rackauckas for visits to three of our node universities. He is: 🔹VP of Modelling & Simulation - #JuliaHub 🔹Director of Scientific Research - #PumasAI 🔹Co-PI - Julia Lab at #MIT 🔹Lead developer - #SciML Open Source Software Org @chrisrackauckas.bsky.social ( 🧵1/2)

122

Chris Rackauckas @chrisrackauckas.bsky.social · 18/09/2025

Check out the JuliaCon workshop SciML in Fluid Dynamics (CFD): Surrogates of Weather Models. Go into depth on all of the major surrogate architectures and give them a try on the weather model challenge problem! www.youtube.com/watch?v=PfRx... #julialang #juliacon #sciml

youtube.com

SciML in Fluid Dynamics (CFD): Surrogates of Weather Models | JuliaCon 2025 | Rackauckas, Abdelrehim

YouTube video by The Julia Programming Language

050

Chris Rackauckas @chrisrackauckas.bsky.social · 17/09/2025

Sundials.jl v5.0: Update to SUNDIALS v7 and Improved DAE Initialization A major update that brings significant improvements to differential-algebraic equation (DAE) solving and upgrades to the latest Sundials C library sciml.ai/news/2025/09... #julialang #sciml #sundials #dae

sciml.ai

SciML: Open Source Software for Scientific Machine Learning

Open Source Software for Scientific Machine Learning

040

Chris Rackauckas @chrisrackauckas.bsky.social · 15/09/2025

SciML Fellowship development 2025 - JumpProcesses.jl: Introducing vr_aggregator for VariableRateJumps, GPU-enhanced SimpleTauLeaping and Extending with τ-Leap Algorithms for Jump Process Simulation #julialang #ssa #gillespie #sciml #tauleap sciml.ai/news/2025/09...

sciml.ai

SciML: Open Source Software for Scientific Machine Learning

Open Source Software for Scientific Machine Learning

040

Chris Rackauckas @chrisrackauckas.bsky.social · 15/09/2025

Introducing SymbolicSMT.jl for symbolic constraint solving and theorem proving! Built on Z3, test the feasibility of symbolic expressions built using Symbolics.jl. Given Constraints([x > 0, y > 0, x^2 + y^2 <= 1]), ask issatisfiable? isprovable? #julialang #symbolics #smt sciml.ai/news/2025/09...

sciml.ai

SciML: Open Source Software for Scientific Machine Learning

Open Source Software for Scientific Machine Learning

0120

Chris Rackauckas @chrisrackauckas.bsky.social · 13/09/2025

New episode of the #julialang Dispatch podcast, featuring Jesper and Philip from Cognibotics, a Swedish robotics company that developed a statically-typed dialect of Julia designed specifically for real-time robot programming! www.youtube.com/watch?v=8wja...

youtube.com

Juliet & Romeo with Jesper Öqvist and Philip Olhager

YouTube video by Julia Dispatch

1120

Chris Rackauckas @chrisrackauckas.bsky.social · 09/09/2025

Yes I'll be going all over Australia. Happy to connect with anyone while there.

061

Reposted by Chris Rackauckas

MACSYS @macsys.bsky.social · 09/09/2025

MACSYS will host Dr Chris Rackauckas for visits to three of our node universities this month. He is: 🔹VP of Modelling & Simulation - #JuliaHub 🔹Director of Scientific Research - #PumasAI 🔹Co-PI - Julia Lab at #MIT 🔹Lead developer - #SciML Open Source Software Org @chrisrackauckas.bsky.social

132

Chris Rackauckas @chrisrackauckas.bsky.social · 07/09/2025

Mixed precision and automatic GPU offloading for Newton solves and stiff ODE solvers has now landed thanks to improvements in LinearSolve.jl. Run LinearSolveAutotune.jl so that things like usage of GPUs in #julialang #sciml becomes automatic! sciml.ai/news/2025/09...

sciml.ai

SciML: Open Source Software for Scientific Machine Learning

Open Source Software for Scientific Machine Learning

2162

Chris Rackauckas @chrisrackauckas.bsky.social · 07/09/2025

New Symbolic-Numeric algorithm which uses rational polynomial interpolation mixed with differential algebra in order to give a highly robust method for parameter estimation and solving inverse problems on ODEs. #julialang #sciml #symbolicnumeric #ode www.sciencedirect.com/science/arti...

sciencedirect.com

Robust parameter estimation for rational ordinary differential equations

We present a new approach for estimating parameters in rational ODE models from given (measured) time series data. In typical existing approaches, an …

062

Chris Rackauckas @chrisrackauckas.bsky.social · 16/08/2025

LinearSolve.jl just got autotuning 🔥 The problem: picking linear solvers is hardware-dependent voodoo. Solution: automatic benchmarking that tells you EXACTLY what to use. Share the results to improve the system! Details: sciml.ai/news/2025/08/16/linearsolve_autotuning/ #julialang #sciml

sciml.ai

SciML: Open Source Software for Scientific Machine Learning

Open Source Software for Scientific Machine Learning

0384

Chris Rackauckas @chrisrackauckas.bsky.social · 11/08/2025

What is the #julialang programming language? This is stitched from the interviews of a lot of developers telling their stories as to why they are committed to the mission of #Julia. Performance, scaling, easy, fun? www.youtube.com/watch?v=jPDC... #sciml #scientificcomputing #hpc #scaling

youtube.com

What is the Julia Programming Language?

YouTube video by JuliaHub

072

Chris Rackauckas @chrisrackauckas.bsky.social · 05/08/2025

#Julialang Dispatch: we welcome Jacob Zelko, who takes us on a fascinating journey from his early days as a #biomedical engineering student through his work at the CDC during the COVID-19 pandemic, and into his current exploration of applied #categorytheory. www.youtube.com/watch?v=HyMp...

youtube.com

Julia Health with Jacob Zelko

YouTube video by Julia Dispatch

081

Chris Rackauckas @chrisrackauckas.bsky.social · 25/07/2025

New paper on fitting #neuroscience models to neuroimaging and electrophysiological data! We show that the previous techniques within the field (Spectral DCM) can be greatly improved using automatic differentiation, #julialang #sciml, ModelingToolkit, and more.

160

Chris Rackauckas @chrisrackauckas.bsky.social · 20/07/2025

How successful has the SciML Small Grants program been at getting newcomers to contribute to open source software #oss for #sciml #julialang and #ai4science? Very! * 13 total projects initiated * ~90% success rate See the blog post summary of the first year! sciml.ai/news/2025/07...

sciml.ai

SciML: Open Source Software for Scientific Machine Learning

Open Source Software for Scientific Machine Learning

0125