Hello Reader!

Ever get stuck on an optimization problem and wish you could get quick help without waiting for forum responses or digging through documentation for hours?

Insights from the Pre-Summer-Break Conference Week

The last week of June was busy in the optimization world. While many practitioners were traveling between conferences — Gurobi's events in Berlin and Amsterdam, plus the academic EURO conference in Leeds — there were some interesting developments worth sharing.

Gurobi Launches the Gurobot

During the busy conference week, Gurobi launched Gurobot, their optimization-focused AI assistant. The origins go back a couple of years. At the Barcelona Summit in 2023, I learned that some technical people at Gurobi had been experimenting with an LLM trained on their documentation. As a test, they used it to answer an actual support query from their forum. After the Gurobi experts team reviewed the response, they decided it was good enough to post as their official answer.

After several rounds of refinement, Gurobot can now answer questions about Gurobi interfaces and even suggest complete optimization models. The integration allows for seamless creation of formal support tickets when you need human help, and it has context awareness about your Gurobi profile and license setup.

How to Use Gurobot Effectively

I tried it myself recently when I needed a simple example model for a client presentation. Instead of getting a textbook reference that would require hours of searching, Gurobi delivered exactly what I needed—a focused example demonstrating the specific issue I wanted to show.

To get good results with Gurobot at gurobi.com (you just need a Gurobi account):

• Be specific in your prompts: "Help me model a production scheduling problem where each machine can only work on one job at a time"
• Provide context about your industry or problem type
• Use detailed prompts — more information leads to better suggestions
• Take advantage of the escalation to human experts when needed

Andreas Wächter Joins Gurobi

Andreas Wächter, the main developer behind the popular open-source nonlinear solver IPOPT, recently joined Gurobi's development team. With many US universities losing attraction for top talent, it makes sense that optimization experts are moving to where cutting-edge development happens. This will likely boost Gurobi's nonlinear optimization capabilities — both for local optimization (which IPOPT handles well) and global optimization, since global techniques use local optimization internally.

Time to Dream

If you had a magic wand, what kind of nonlinear problem would you love to see solved instantly? Hit reply and let me know — I'll also share what my first gut instinct told me.

Until the next iteration!

Tim Varelmann

​Bluebird Optimization​

Complicated Decisions - Simply Automated!

Follow me on LinkedIn​

P.S. If you try Gurobot, I'd be curious to hear how it compares to your usual approach for getting optimization help.