AI & External Data Applications

Understand what AI really changes for your business, then build a working AI app from scratch, right in the lecture room.

Duration: 1 day (interactive)
Delivery: On-site or online
Group size: Up to ~15
Languages: Czech or English
Prerequisites: None, no coding required
Investment: Quote on request

Who this is for

AI is everywhere; clarity is rare. This workshop is built for a general, non-technical audience: first we separate real business value from hype in plain language, then we get hands-on and turn external web data into a working AI product with current tools. It's for managers, founders, marketing and operations, anyone who wants to use AI well. No coding required.

Key takeaways

Explain what today's AI can and can't do, in plain terms.
Spot AI use cases, including simple automation, that pay off in your work.
Prompt effectively (zero-shot, few-shot, chain-of-thought).
Get oriented: model providers, AI coding IDEs, and what AI coding actually costs.
Use AI responsibly and protect data.
Build and ship a small AI app that uses live external data.

What we cover

01
AI context, in plain language
What today's AI is, how language models work (without the jargon), and where the real value sits vs. the hype.
02
Prompting that works
Anatomy of an effective prompt; zero-shot, few-shot and chain-of-thought, with practice.
03
Providers, IDEs, cost & automation
Who provides the models; what an AI coding IDE is and which to use; what coding with AI actually costs; the basics of automating a task, plus responsible use, privacy and hallucinations.
04
Build lab, part 1: idea → data
Frame a small project and pull external web data with Exa, assisted by Claude / Claude Code.
05
Build lab, part 2: app → live
Reason over the data with an LLM and deploy a working app to the web with Vercel.

Inside this track

Effective AI in Business

A no-prerequisite primer for a general audience: spot the use cases that pay off, prompt so it actually works, and navigate the flood of AI tools, responsibly.

Build an AI Project from Scratch

An interactive lab: go from idea to a live app in the room using Claude, Claude Code, Exa (web data) and Vercel (deploy). External data → AI → a finished product.

Tools we use

AI coding IDEs (Claude Code, Cursor, VS Code)Claude (build assistant)Exa (external web data)Vercel (one-click deploy)Model providers (Anthropic, OpenAI, Google)Cost & automation basics

How it's taught

A short primer, then a live build: concepts up front, then most of the day spent making something real. Everyone leaves with a working demo and the confidence to repeat it.

About your lecturer

Jan Černý

Researcher & consultant, Competitive & Technology Intelligence

Jan Černý holds a Ph.D. in applied informatics (Prague University of Economics and Business) and a PhDr. in information science (Charles University). He has spent fifteen-plus years advising strategy, innovation and IP teams on competitive and technology intelligence, technology scouting and applied AI, and researches the same subjects academically, publishing in venues such as The Electronic Library, Review of Managerial Science, the Journal of Information Warfare and Online Information Review.

See research & publications

Ph.D., Applied Informatics, Prague University of Economics and Business
PhDr., Information Science, Charles University
15+ years advising strategy, innovation and IP teams
Peer-reviewed author in CI, OSINT, patent analytics and applied AI
Delivered with the National Library and the Czech and Slovak Ministries of the Interior

Frequently asked questions

Do I need technical skills or coding?

No. The track is designed for a general business audience. The build lab uses AI-assisted tools (Claude Code, Vercel) so that non-coders can ship a working application by the end of the day.

What will we actually build?

A small but real application that pulls external web data (via Exa), reasons over it with a large language model (Claude), and is deployed live to the internet (Vercel), assembled with the help of Claude Code. External data → AI → a finished product.

Is 'prompt engineering' still a real skill?

Yes. Structured prompting, zero-shot, few-shot and chain-of-thought, measurably improves output quality, and the skill transfers across models and tasks (Černý, Journal of Information Warfare, 2024; Černý, 'AI-Driven Generation of Key Intelligence Topics and Questions', ECKM 2025).

Can AI be trusted for business decisions?

Treat AI as an accelerator, not an oracle: outputs must be verifiable and reviewed. Well-designed agentic AI systems can reliably extract value from large external data sets when built with checks and a clear chain of evidence (Černý, Avramov & Pendse, The Electronic Library, 2026, DOI 10.1108/EL-06-2025-0272).

What about hallucinations, privacy and ethics?

We cover where models predictably fail, how to manage hallucination, and the basics of data protection and responsible use, hands-on, as part of choosing and using tools.

What tools do we use, and what does AI coding cost?

We work with the main model providers (Anthropic, OpenAI, Google) and AI coding IDEs: chiefly Claude and Claude Code, plus Exa for web data and Vercel to deploy. I'll explain what an IDE is, which ones suit non-coders, and what a small AI build actually costs to run, from free tiers to usage-based API pricing, so there are no billing surprises.

Sources & further reading

Tell me which track fits and a bit about your team, and I'll reply within 48 hours with formats, dates and a tailored quote.