From Ambiguous Problems to In-depth Analysis - Valuable Insights from Manus.im

In the workplace, have you ever encountered a situation like this?

Your boss casually throws out an ambiguous request, such as: "Help me hire a reinforcement learning engineer."

At this point, a common reaction from beginners is:

• They start by looking at candidates' resumes, checking QS rankings, education, work experience...

• Then they begin to struggle with what criteria to use for ranking, getting more confused and directionless...

• Finally, after spending a lot of time on a report, the boss glances at it and says, "You missed the point."

😨 A lot of effort wasted, and efficiency is poor!

The expert approach is to use logical thinking to break down the task and turn a vague problem into a structured analysis.

Step 1: Break Down the Task and Identify the Real "Key Criteria"

When faced with a task, you shouldn't make decisions based on intuition. Instead, you need to stand in the boss's shoes and understand what they truly want.

✅ The boss wants the candidate's real reinforcement learning ability, not QS rankings or high education.

So, we need to first define key criteria, such as:

1. Project Experience — Has this person actually worked on reinforcement learning projects?

2. Code Quality — Is the code open source? Are there any real contributions? Can the code run?

3. Past Contributions — Do they have papers, competition awards, or industry recognition in reinforcement learning?

These criteria allow decisions to be more data-driven rather than based on subjective judgment.

---

Step 2: Data-Driven — Use Clear Data Analysis to Make Reliable Conclusions

With the criteria set, the next step is to quantify the data and put the analysis into practice.

📊 How to avoid "feeling good"? Use data scoring!

This way, you can clearly tell the boss:
"We scored candidates based on reinforcement learning project experience, code quality, and contributions. The highest score is C, so C is more suitable."

✅ This way, the boss only needs to glance at the conclusion to make a quick decision without being overwhelmed by a lot of scattered information.

---

Step 3: Efficient Workplace Habits — Record Everything to Make Tasks Trackable

Here, the "Todo List" approach in Manus Demo is very insightful.

Many workers are accustomed to using Excel to record what they do every day, including time, tasks, and progress...

This not only helps them review their work but also creates a clear task trail for easy review and optimization.

🎯 Three functions of a work log:

1. Prevent forgetting task details

2. Show your progress to the boss, increasing trust

3. Enhance a sense of accomplishment by seeing how much you've completed!

Have you ever had such an experience? Worked all day but can't recall what you actually did...
At this time, if you have a clear Todo List, your daily efforts can be quantified, and growth becomes visible!

---

Step 4: How to Write Emails/Reports That the Boss Can Understand at a Glance?

💡 The boss doesn't have time for long essays, so reports should have these three points:

1. Summary at the top — Let the boss see the most critical insights at a glance.

2. Logically break down the vague problem — Tell the boss how you derived the conclusion.

3. Data supports decision-making — Use data, not "I think," to persuade the boss.

Example 1 (Wrong Demonstration):

🚫 Report the boss can't understand:

These 10 candidates graduated from Stanford, MIT, CMU... Their QS rankings are 1, 3, 5... We ranked them based on their educational background as follows...

👎 Problem: The boss doesn't care about QS rankings; they want to know who has stronger reinforcement learning skills!

Example 2 (Correct Demonstration):

✅ Report the boss prefers:

Summary:
We scored 10 candidates based on reinforcement learning project experience, code quality, and paper contributions. The highest score is C (total score 13), recommended for priority consideration.

Analysis Process:

• We focused on candidates' practical experience in reinforcement learning, not education.

• C participated in 2 reinforcement learning projects, has excellent code quality (with open source contributions), and published papers, thus ranking first.

• A and B have advantages in different areas, with the same total score, but slightly weaker code ability than C.

Data Analysis: (Attached Excel sheet)

👑 Such a report allows the boss to understand your thought process at a glance, making decisions more efficient!

---

Step 5: Critical Thinking — How to Demonstrate Thinking Ability in Daily Work?

In Manus's second Demo, there is an example about Redfin vs. Trulia.

In the U.S. real estate market, some know Redfin, and some know Trulia, but their user groups and market positioning are slightly different.

This actually involves a very important question:
🤔 When faced with multiple choices, how do you make decisions?

🎯 Excellent professionals actively think about the logic behind information rather than passively accepting information.

Just like YZ said, "The new team member looked at his report and asked what to do."
True experts won't wait for others to tell them what to do but will actively think about "why do it?" and "how to do it better?".

---

Summary: How to Turn Vague Problems into High-Quality Analysis?

💡 Summary of "Expert Thinking" in the Workplace:
✅ Break down tasks and grasp the real key points (Logical Thinking) — Don't get stuck in vague problems, but break them down into structured analysis.
✅ Speak with data, not intuition (Data Analysis) — Make decisions more evidence-based, not just gut feelings.
✅ Efficient recording, enhance a sense of accomplishment (Todo List habit) — Make work organized, easy to review and optimize.
✅ Write reports for the boss, starting with conclusions, then logic — Let the boss see key insights at a glance, improving work efficiency.
✅ Cultivate Critical Thinking and actively think about the logic behind it — Enhance your analytical skills, not just passive execution.

If you can master these ways of thinking, whether in the workplace or in future entrepreneurship, you can solve problems more efficiently!