[08]
AI TRAININGS FOR ASSECO
█ TIER A · YEAR 2026 · STATUS: LIVE
Three authored AI courses for a corporation — from the board to tech teams
[FIG. 1] MISSION
My own AI trainings delivered for Asseco Data Systems — three full workshop courses grounded in real business cases, not theory. (1) „AI as strategic management support” (8h for directors/managers): 7 cases from capacity planning through steering-committee pre-reads and tender/SIWZ analysis to sprint-metric interpretation; e.g. cutting decision meetings from an hour to 15 minutes. (2) „Prompt Engineering & Custom GPT” (8h): 10 cases — system-prompt design, building Custom GPTs / Gemini Gems, RAG in Notebook LM, local models (LM Studio) on sensitive data, methodical testing (benchmark, adversarial, rubrics). (3) „AI in VS Code”: requirement extraction and user stories from procurement (SWZ) documents. Each course is a coherent package: context, a trainer version with ready prompts, cases and slide decks.
[FIG. 2] ARCHITECTURE
hover a block to see its description
[FIG. 2A] WORKSHOP PROGRAM
STEERING COMMITTEE: ▓▓▓▓▓▓▓▓▓▓▓▓ 60 MIN
| INPUT | SYSTEM PROMPT | OUTCOME |
|---|---|---|
| DAILY NOTES | [role + context + recommendations] | decisions by morning |
| JIRA REPORT | [metrics → business narrative] | project status in plain words |
| STEERING COMMITTEE | [decision pre-read] | 60 → 15 min |
[FIG. 3] CHALLENGES
[+][CH-01]
Real cases instead of theory — each module built on a concrete situation (SIWZ, JIRA, steering committee) with a ready input and output.
[+][CH-02]
A non-technical audience — translating LLMs into a decision-maker's language, no jargon, no coding prerequisites.
[+][CH-03]
Sensitive data — a dedicated track on local models (LM Studio) for data that can't go to the cloud.
[FIG. 4] AI LAYER
Three authored AI courses for a corporation — from the board to tech teams
[FIG. 5] GALLERY
SCREENSHOTS INCOMING — see SHOTS-WANTED