AIatWork

Track changes shows you where something changed.
It doesn't tell you why it matters.

Long documents (contracts, policies, proposals, multi-version reports) are the quiet productivity tax most teams pay every week. Hours of line-by-line review, looking for what actually changed and what it means for the business.

Tonight's session with Rui Vas, 3x startup founder, AI educator, and co-founder of SuperOperator.ai, with a decade at the edge of AI (from building robots for Swiss Pharma labs to leading AI strategy for Ubuntu Linux with Google, Microsoft, and Nvidia), shows how to use AI to actually understand long documents, not just read them faster.

The focus is not speed. The focus is clarity: knowing what changed, why it matters, and what to do about it.

In this session you'll learn how to:
🔷 compare long documents and identify the differences that actually matter
🔷 track changes and understand what changed, not just where
🔷 extract key insights, risks, and important updates from dense source material
🔷 summarize complex documents into clear, decision-ready outputs
🔷 use AI to reduce manual reading without missing what matters

Walk away ready to handle your next contract revision, policy update, or 80-page report without scrolling line by line.

👉 Step inside AI at Work: https://aiatwork.ai/onboarding/begin

See you tonight.

Claude can write code.
Claude can analyse 100-page documents.
Now Claude can design.

Anthropic released Claude Design, a tool that lets non-designers go from a rough idea to a polished, on-brand visual deliverable through plain conversation. No Figma. No design background. No designer briefing required.

Tonight's session with Álvaro Lamas Fuente, Professor of Generative AI at EBIS Business Techschool and former Customer Solutions Engineer at Google, runs four live builds in Claude Design, each more complex than the last: a brand design system, a one-pager, a pitch deck from a rough outline, and a landing page that gets handed off directly to Claude Code for implementation.

In this session you'll learn how to:
🔷 set up a Design System once by pointing Claude at your brand reference, so every asset is automatically on-brand
🔷 apply the Describe → Generate → Refine → Export framework to any visual task
🔷 build a one-pager, pitch deck, and landing page using your Design System as the visual foundation
🔷 refine AI-generated visuals using three editing modes (chat, inline comments, direct edits)
🔷 export finished designs in the right format for each use case, including a direct handoff to Claude Code for implementation

Walk away with a repeatable process for going from rough idea to shareable, on-brand visual in minutes, not days.

👉 Step inside AI at Work: https://aiatwork.ai/onboarding/begin

See you tonight

May at AI at Work brought together mentors from Microsoft, IBM, Novo Nordisk, Google, Tesla, BlueAlpha, EBIS Business Techschool, and consultants who've worked with brands like Estée Lauder, NIVEA, Forbes, and the European Parliament.

10 sessions. 10 real problems. Here's what we learned together:

🔷 how to use AI to model the financial impact of sales and pricing decisions
🔷 how to set up ChatGPT and Claude for professional use, with custom instructions, projects, and four prompting techniques
🔷 how to map capabilities, surface skill gaps, and model workforce cost scenarios for strategic talent planning
🔷 how to turn customer feedback (transcripts, tickets, surveys) into structured, actionable insight
🔷 how to load structured data, train simple ML models, and build repeatable pipelines in Python
🔷 how to design UX systems from user flow to interface with AI as a thinking partner
🔷 how to clean, structure, and explain messy data inside Excel with AI
🔷 how to build working apps with Lovable, no developer required
🔷 how to prepare feedback, simulate difficult conversations, and support team development with AI
🔷 how to design, document, and automate full departmental workflows end to end

Every session was built around one rule: walk away ready to apply it the next working day.

Same tempo in June. New mentors, new workflows, new tools.

The June lineup drops in a few days.

👉 Step inside AI at Work: https://aiatwork.ai/onboarding/begin or join through the link in our bio

Most teams don't have broken processes. They have undocumented ones.

The onboarding steps live in someone's head. The approval thresholds depend on one person remembering. The handoff between Finance and Legal works… until the person who held it together is on holiday.

That's when you find out which of your workflows were actually workflows, and which were just habits.

Tonight's session with Álvaro Lamas Fuente, Professor of Generative AI at EBIS Business Techschool and former Customer Solutions Engineer at Google, shows how to use AI to design, document, and automate the workflows your team runs every day. No process consultant, no diagramming tool, no technical background required.

Two live demos walk through real cases (HR applicant screening and vendor onboarding in Operations), mapping each process, auditing it for failure risks, and turning it into a business-ready document. The vendor workflow then gets automated live in Make, so you see what it looks like when a process stops depending on someone remembering to act.

In this session you'll learn how to:
🔷 apply a five-component framework (Trigger, Steps, Decision Points, Outputs, Owner) to any real process
🔷 use Claude or ChatGPT to map a team workflow from plain language into a structured, audited document
🔷 choose between three output formats (checklist, SOP, or scorecard) based on what the team actually needs
🔷 identify which steps belong to humans and which are real candidates for automation
🔷 build a no-code automation in Make that triggers on a workflow event and routes based on defined logic

Walk away ready to book 30 minutes this week and map one real process, before involving the team.

👉 Get unified access to ChatGPT, Claude, Gemini and live mentor sessions: https://aiatwork.ai/onboarding/begin

See you tonight.

Every manager has a conversation they've been postponing.

The performance issue nobody wants to name. The feedback that's been rehearsed in the shower for two weeks. The team development decision that's easier to defer than to deliver.

People management is mostly the conversations - and most of us were never trained for them.

Tonight's session with Alicia Carabias — Talent Acquisition Leader at IBM, with 25+ years across PwC and IBM leading international teams — shows how AI can act as a thinking partner for the hardest parts of leadership: structuring feedback, preparing difficult conversations, and making team development decisions with more clarity and less anxiety.

Not to replace empathy or judgment — but to sharpen them.

In this session you'll learn how to:
🔷 turn unstructured notes and observations into clear, actionable feedback
🔷 prepare for difficult conversations by simulating scenarios before they happen
🔷 adapt tone and framing for different team members and situations
🔷 use AI to support — not outsource — team development decisions
🔷 understand the limits: where empathy, emotional intelligence, and human judgment must lead

Walk away with frameworks you can use the next time a 1:1 needs to go somewhere harder than usual.

👉 Step inside AI at Work: https://aiatwork.ai/onboarding/begin

See you tonight.

300+ companies at Macedonia2025 proved it: the real gap isn't between AI adopters and non-adopters. It's between those treating AI as optional and those using it to work 3x faster.

Our mentor Daisy Ilaria Illaria, ex-Philips, ex-PVH Corp., showed how teams using AI aren't just keeping up. They're setting the pace.

The data doesn't lie: AI-driven teams hit targets 2.5x more often.
Think upskilling is complex? Our platform gets teams productive in under 10 hours.

Worried about ROI? Most members 3x their output within the first year.

The risk isn't moving too fast, it's moving too slow.

Start your AI journey today with zero risk:
👉 https://lnkd.in/evHAwsqa

Most data work stops at the spreadsheet.

You can sort it, filter it, pivot it, but the moment someone says "build a model from this" or "automate this end-to-end," the wall goes up.

The gap between analysing data and shipping a working pipeline is where most people stall.

Tonight's session with Clara Asensio Martinez, Data & AI Technical Trainer at Microsoft, with 10+ years in cloud and AI, walks through that full path: from raw tables, to trained models, to repeatable pipelines.

Live Python demos in VS Code with Copilot show how the ideas translate into real code and real results. No prior ML background required.

In this session you'll learn how to:
🔷 define what structured data actually is and why schema matters
🔷 load structured data into a table and run clean transformations
🔷 train a simple machine learning model on tabular features
🔷 understand how preprocessing shapes what a model can learn
🔷 build a repeatable data pipeline that connects preparation, training, and evaluation in one automated workflow

Walk away ready to take raw data, prepare it properly, and run it through your first end-to-end pipeline the next working day.

👉 Want in on tonight's session?
Join here: https://aiatwork.ai/onboarding/begin

See you tonight.

Customer feedback piles up faster than anyone can read it.

Interview transcripts. Support tickets. Survey responses. Lost-deal notes. Most of it sits unread in folders and inboxes - while decisions still get made on gut feel.

The patterns are there. Nobody has time to find them.

Tonight's session with Matthias Stepancich - Growth at BlueAlpha, Tesla's first Growth Performance Manager, and the marketer who scaled paid media from zero to $10M+ per quarter — shows how to turn unread feedback into structured, actionable insight using AI.

Five live demonstrations in Claude walk through real workflows: analyzing interview transcripts, synthesizing patterns across feedback sources, building customer health scores from behavioral signals, and generating custom discovery questionnaires from a single methodology document.

In this session you'll learn how to:
🔷 transform unstructured feedback (transcripts, tickets, surveys) into categorized, severity-tagged analysis
🔷 synthesize multiple feedback sources to surface cross-cutting patterns invisible in any single one
🔷 set up a Claude Project with persistent context that eliminates repetitive setup across sessions and teammates
🔷 design a customer health scoring framework based on behavioral signals, not lagging metrics
🔷 build a methodology reference guide that turns Claude into a trained specialist for your team

Walk away with a downloadable toolkit: question bank, prompt library, fillable insight-to-action spreadsheet, and ready-to-use Claude Project instructions.

👉 New here? Start your AI at Work journey: https://aiatwork.ai/onboarding/begin

See you tonight.

Workforce planning still happens in scattered spreadsheets, disconnected HR systems, and gut-feel meetings.

Critical talent decisions get made with half a picture. Skill gaps surface too late. Succession plans live in slides nobody updates.

Tonight's session with Sebastian Sbirna — Data Governance Lead at Novo Nordisk, with 8+ years driving data and AI initiatives across pharma, banking, and energy — shows how AI becomes a "thinking companion" that synthesizes fragmented workforce data into a coherent strategic picture.

Through four live demonstrations in Claude, you'll see how AI can map capabilities, surface hidden talent patterns, and model cost scenarios in real time.

In this session you'll learn how to:
🔷 identify the core challenges in traditional workforce planning that AI can actually solve
🔷 run a capability mapping exercise to reveal hidden talent patterns in your team
🔷 analyse strategic skill gaps by comparing current data against future business needs
🔷 model workforce cost scenarios to support joint HR and Finance decisions
🔷 implement a framework for responsible AI use — privacy, governance, and output verification

Walk away ready to run your first Capability Audit the very next day.

👉 Not inside AI at Work yet? Join here: https://aiatwork.ai/onboarding/begin

See you tonight.

Most people use ChatGPT and Claude the same way they did the day they signed up.
No custom instructions. No privacy settings reviewed. No clear logic for which tool to open when.

The result: inconsistent outputs, the same context re-explained in every new chat, and a vague sense that AI "kind of works."

Tonight's session with Beatrix Mészáros - AI Marketing Consultant with 15+ years across 100+ global brands - focuses on turning casual AI use into a structured, professional setup you can rely on daily.

Built around a single real scenario, you'll see the same task run through both tools and four different prompting techniques, side by side.

In this session you'll learn how to:
🔷 configure custom instructions in ChatGPT and personal preferences in Claude
🔷 manage privacy and training controls in both tools
🔷 apply four prompting techniques (zero-shot, role+goal, few-shot, chain-of-thought) to the same task
🔷 organise ongoing work with Custom GPTs and Claude Projects
🔷 decide when to reach for ChatGPT and when to reach for Claude

Walk away with a reusable setup, a four-technique prompt library, and a clear decision framework.

See you tonight inside AI at Work.
Join here: 👉 https://aiatwork.ai/onboarding/begin