Real Trainee’s Takeaways: How AI Training with Deepseek Actually Changed the mindset

Real Trainee’s Takeaways: How AI Training with Deepseek Actually Changed the mindset

Overview

On June 18, 2025, System in Motion hosted its latest session of AI Fundamentals with DeepSeek (L1 & L2) at the CCI FRANCE CHINE offices in Shanghai. Presented by Elodie Abbé, a seasoned executive coach and talent development expert, the 3-hour workshop was designed to empower professionals with the tools and understanding needed to harness Generative AI in real business settings.

Introduction & Context

Participants introduced themselves, their roles, their familiarity with Generative AI, and the last thing they had used AI for. This warm, inclusive kickoff set the tone for an engaging and practical learning experience. Elodie Abbé, the lead trainer, brought 25 years of cross-cultural talent development experience to the room.

“The most valuable asset of a 21st-century institution will be its knowledge workers and their productivity.” This session was all about enabling productivity through smart AI integration—moving from abstract understanding to concrete, hands-on experience, exploring two major type of models, two additional brains:

The DeepSeek Platform & Training Journey

Participants explored System in Motion’s AI platform, Ai-in-Motion(AIM) which gives users access to a suite of cutting-edge AI tools such as GPT 4o, deepseek V3 & R1, kimi and etc,. It served as the central hub for the day’s exercises, offering both DeepSeek V3 and R1, among other top-tier models like GPT-4o. The course itself was structured into five chapters, each building upon the last:

Chapter 1: Fundamentals of Generative AI

The day began with a concise overview of core GenAI principles—its capabilities, use cases, and the foundation of prompt-based interaction. The key message: AI is only as good as your instructions. “Use natural language to give orders to the machine” became an ongoing refrain, reminding participants to treat AI more like a partner than a black box.

Chapter 2: Level 1 Prompting with ERCFC

Elodie introduced the ERCFC Framework, a structured method to craft high-impact prompts:

• E – Role: Who is the AI acting as?
• R – Context: What background should the AI know?
• C – Format: What structure should the output follow?
• F – Command: What exactly should the AI do?
• C – Clarity: Are the instructions unambiguous? Live examples highlighted the difference between vague queries and precise instructions. One key insight: always refresh a new conversation for focused prompts—context pollution can lead to unpredictable results.

Chapter 3: Corporate Implementation Strategies

The training then shifted to business use cases. Participants examined how GenAI can be embedded into company workflows for tangible ROI. A striking chart illustrated that efficiency (measured in intelligence per cost) often matters more than raw performance, underscoring the importance of choosing the right model, not just the most popular.

Another key takeaway: never rely on a single model. The strength of GenAI lies in flexibility. The session emphasized using multiple LLMs (Large Language Models) for comparison and optimization.

Chapter 4: Advanced Reasoning Models – Chain of Thought

Moving into more advanced territory, Elodie introduced reasoning models like DeepSeek R1 and Claude 3.7. Attendees practiced “0-shot,” “1-shot,” and “few-shot” prompting techniques while learning how Chain of Thought can lead to stronger, more logical AI responses. Hands-on exercises allowed participants to give the same prompt to different models and compare the results in real time—offering immediate insight into each model’s strengths.

Chapter 5: Business Applications & Knowledge Management

The session wrapped with real-world implementation strategies, focusing on knowledge management integration. Participants discussed how to connect internal business data with text models to improve AI reliability, strategic analysis, and internal knowledge workflows. This segment brought theory into tangible business relevance, sparking discussions around data governance, shadow AI, and one-shot AI in corporate environments.

Final Thoughts: Training for the Future

System in Motion’s AI Fundamentals with DeepSeek training proved that empowering knowledge workers goes beyond explaining concepts—it’s about enabling action. From practical prompting strategies to advanced reasoning exercises, participants left the session equipped with the tools to make GenAI work for them.

Want to empower your team or yourself with GenAI skills that drive real impact?

We are Here to Empower

At System in Motion, we are on a mission to empower as many knowledge workers as possible. To start or continue your GenAI journey.