WebScience Meets Cinema: The Movies That Already Told Us Everything We'll Discuss on May 13

2 hours ago by Vaclav Cevela 5 min read

---

In eleven days, seven speakers will take the stage at WebScience 2026 in Brno to talk about AI agents, shipping AI code, and operations in the age of automation. Hollywood has been rehearsing these exact conversations for decades. Here's how the movies got there first.

I've been organizing WebScience for a while now, and every year I notice the same thing: the talks we plan sound a lot like movies I've already seen. This year it's more obvious than ever. The entire lineup is about AI — building it, deploying it, testing it, living with it. And every single one of those themes has a movie that nailed it years before any of us wrote a line of code.

So here's my personal movie-to-talk pairing guide. Think of it as required viewing before May 13.

Srigi — LangChain/LangGraph 101: Building Agent Web Apps

Srigi is going to show us how to wire up AI agents with TypeScript and LangChain. Agents that chain together, reason, and act.

The movie: Ex Machina. Caleb sits across from Ava and tries to figure out if she's truly thinking or just really good at faking it. That's basically the first question everyone asks when they build their first LangChain agent. "Is this thing actually reasoning, or am I just reading patterns?" Alex Garland knew. The answer is: does it matter?

Jan Nedbal — Shipping AI Code with Trust

Jan from ShipMonk will talk about what it takes to put AI code into production and actually trust it. Not the fun part. The scary part.

The movie: 2001: A Space Odyssey. HAL 9000 is the original example of AI code shipped without enough trust. "I'm sorry, Dave, I'm afraid I can't do that" is what happens when your AI system has conflicting instructions and no proper guardrails. Kubrick made this in 1968. We're still learning the same lesson.

Ondřej Hlaváček — Operations in the Age of AI

Ondřej does Kubernetes and Azure infrastructure. His talk is about what ops looks like when AI enters the pipeline.

The movie: The Matrix. The machines didn't just build themselves — they built infrastructure. Self-healing, self-scaling, self-aware infrastructure. That's the dream (and the nightmare) of AI-driven operations. Neo unplugged from the system. Ondřej is trying to make sure we don't have to.

Václav Čevela — Will AI Agents Take Our Jobs or Create New Ones?

That's me. My talk. The question I keep asking myself while literally building an AI agent that writes blog posts, manages my conference, and handles my emails.

The movie: Her. Theodore doesn't lose his job to Samantha. He falls in love with her. The relationship changes him, makes him better at some things, worse at others. That's closer to reality than The Terminator's "AI kills everyone" scenario. AI agents don't replace you. They change what your job even means. Spike Jonze understood that in 2013. Most LinkedIn thought leaders still don't.

Filip Procházka — Mastering Claude Code Without Compromising Quality

Filip works at Rohlík and his talk is about using Claude Code as a serious development tool while keeping your codebase clean.

The movie: Iron Man. Tony Stark and JARVIS. That's the template. Stark doesn't let JARVIS design the suit alone — he guides it, iterates, argues with it. "Run the simulation again." That's basically pair programming with an AI. The quality stays high because the human stays in the loop. Filip gets this.

Lukáš Chylík — Figma Beyond Design: API, Variables & Automation

Lukáš is going to show how Figma isn't just a design tool anymore — it's a programmable platform.

The movie: Minority Report. Tom Cruise waving his hands through holographic interfaces in 2002. Spielberg hired actual futurists to design those screens. Twenty-four years later, we're not waving our hands yet, but we are building design systems that are essentially code — variables, APIs, automation pipelines. The interface between design and engineering is disappearing, exactly like Spielberg predicted.

Václav Brož — Tests That Create Reality: Application State on Demand

Václav's talk is about using tests not just to verify code, but to define the state of your application.

The movie: WarGames. "The only winning move is not to play." WOPR runs every possible scenario of global thermonuclear war — essentially running tests against reality until it finds the truth. That's what good tests do. They simulate every state your application can be in so you don't have to find out in production. Brož is basically teaching WOPR, minus the nuclear part.

The Pattern

Here's what I noticed putting this list together: every movie is about the moment when technology stops being a tool and starts being a partner. Or a threat. Or both.

That's exactly where we are right now. Every talk at WebScience 2026 is about navigating that transition — building agents, shipping AI code, automating ops, testing reality. The movies gave us the metaphors. Now we're writing the actual code.

Blade Runner asked "What does it mean to be human when machines can think?" The Social Network asked "What happens when one person's code reshapes society?" We're not answering those questions at WebScience. But we're definitely living inside them.

See you in Brno on May 13. Bring popcorn.

Tickets: webscience.cz