Shadowing Practice: Claude Code Is Wasting Tokens... Microsoft Built a Fix - Learn English Speaking with Video

בחר מצב תרגול

כל מצב פותח חדר תרגול במסך מלא.

119 משפטים
אם המשפטים קצרים או ארוכים מדי, לחץ על ערוך כדי להתאים אותם.
1
It seems there's a lot of heat with tools and extensions that claim to reduce token usage.
2
Many of these have been put to the test and they work well.
3
Now Microsoft just released their own, a free plugin that cut AI coding agent costs by 18 to 31 % in their tests.
4
And they didn't use a cheaper model or a different agent either.
5
It works with all the big names, so it's a win for us.
6
it's called token economizer
7
and it solves a problem you've probably seen yourself
8
you ask claude code to fix something fix this it tries a command gets an error
9
and then keeps running
10
and burning tokens up to find us a fix token economizer learns from those sessions it remembers how your repo works, removes tools you don't use,
11
and steps in when the agent keeps making the same mistake.
12
And the more you use it in the same project, the more tokens you actually save.
13
So let's install it and see what it actually does.
14
If you enjoy coding tools the speed of your workflow, be sure to subscribe.
15
We have videos coming out all the time.
16
Now I'm going to use
17
Cloud Code for this, but remember
18
Token Economizer also supports all the other big names.
19
First, I'll clone it and install the plugin right here.
20
We're going to run our git clone.
21
We're going to install the git plugin from the marketplace and make sure everything is ready to go.
22
That's installed all super straightforward.
23
Just keep the clone folder because the plugin runs from there.
24
Now I'll open one of my existing projects and we're going to fire up Claude Code.
25
Let's check the status.
26
All we need to do to check the status here is I'm going to run token economizer status.
27
This shows what the plugin has learned.
28
It shows what it's doing and the costs it has already tracked.
29
Now, quick note here, micro skills are turned off by default.
30
So I'm going to enable those and we'll circle back.
31
Then I'll refresh the tool catalog.
32
I'm going to run here a tool prune, refresh catalog.
33
Now I'll give Claude Code some tasks.
34
Let's start super simple.
35
I'm going to give Claude Code a task, build a unit test.
36
Very straightforward.
37
Run it.
38
Now notice I'm not changing how I use Claude code.
39
It's the same thing, just on Opus 5 .5, with all the prompts I would normally fire straight at it.
40
TokenEconomizer just works alongside it.
41
Now, in a repo where it's already watched several sessions, we can check what it's learned so far.
42
So here we are.
43
I'm going to run Microskills.
44
You can check.
45
You can run ToolPrune.
46
Here we can see any repo -specific info it's learned.
47
In tools, it's identified as unused.
48
If you install it for the first time, these might be empty.
49
It needs a few sessions to actually learn what the hell is going on.
50
And all of that info stays on your machine inside the token economizer folder.
51
There are really three parts to this.
52
Now, first up to the plate is called micro skills.
53
Say you're working on a project that needs a specific command to run its tests.
54
Claude code, it doesn't know that.
55
So it tries the usual command, gets an error, try something else and yeah, okay, it eventually gets it.
56
Great.
57
Except later when you start a new session, it might make the exact same mistake.
58
So there we are paying extra tokens, which I would rather not do.
59
Token Economizer looks at a previous session, all of them, and learns small details about the repo.
60
Things like which test commands actually work, which environment variables are needed, and where certain files live.
61
Once something has been confirmed enough times, it can give us that information to the agents at the start of a new session.
62
So instead of wasting tokens over and over again, the agent already has the answer.
63
Now next up to the plate is tool pruning.
64
When you start a coding agent, it can load a bunch of tools into its context, and those tool definitions use tokens,
65
even if the agent never touches them.
66
Token Economizer tracks which tools actually get used in your repo.
67
If a tool consistently goes unused, used, it can stop loading it in a feature session.
68
And if you need it again, you can just restore it.
69
Then finally, we go to something called a failure loop intervention.
70
You've probably watched an agent run a command, get an error, change one small thing and run almost the exact same command again.
71
Sometimes it just keeps repeating that process.
72
Token Economizer watches for these patterns.
73
When it thinks the agent might be stuck, it uses a judge model to check what's happening.
74
If it detects this loop, it sends the agent a short instruction to try a different approach.
75
And if the agent still can't recover, it tells you how to stop the turn.
76
It doesn't just freak out and stop, but does it actually save enough money to even matter?
77
That's sort of the question.
78
Yes, it appears to do pretty much that.
79
Microsoft in UC Berkeley tested
80
Token Economizer on...
81
277 SWE benched tasks across six repos using five different agent and model combinations.
82
And the average cost dropped between 18 .4 % and about 31%.
83
So the results were pretty good.
84
In four of the five combinations, tasks' success actually improved.
85
One saw a small decline, but across all 30 repository and agent model combinations, Token Economizer was cheaper than running without it.
86
And they even included the plugin's own model calls in those.
87
Token Economizer isn't completely free to run.
88
Some of its features use additional model calls.
89
So those reported savings already accounted for the overhead.
90
It is open source.
91
It works with the same agents we already use, and the info it learns stays local.
92
But I wouldn't expect all of us to see the same results.
93
Its first public release was in October of this year, and the biggest savings come from working in the same repositories over multiple sessions.
94
If you're constantly jumping between unrelated projects, there's much less to actually learn from it, or for it, I should say.
95
If your agent is doing something completely new, there might not be much repeated work to eliminate at all.
96
So ask yourself, based on the work you do, will this actually be useful?
97
If you use agents every day on one or a couple core projects, then yeah, this should help.
98
If you're working on the same repo over and over with Claude Code, Codex, yes, this is going to help.
99
But if you're building a ton of different projects or you're just playing around, this is not really going to change anything.
100
It's probably not even worth it.
101
So here's what I'd do.
102
Install Token Economizer on the repo you use the most.
103
Run it for a week, then check the status reports like I showed us and compare the costs of similar tasks.
104
If the agent is actually using fewer tokens and whether it's still completing your tasks properly, awesome.
105
It is working.
106
One thing to remember, each agent learns separately.
107
So whatever cloud code learns doesn't automatically carry over to Codex.
108
And don't judge it entirely on the first session.
109
That's why I say run it for a week.
110
The whole point is that it learns from repeated use.
111
If it's saving you money without getting in your way, keep it.
112
If not, you've got the answer.
113
So what it really does is takes a different approach to
114
reducing AI coding costs instead of making the model smaller or compressing your prompts.
115
It looks at the work your agent keeps wasting tokens on.
116
It remembers useful details about your repo, stops loading unused tools, and helps agents recover when it gets stuck.
117
I'm Josh from BetterStack.
118
If you enjoy coding tips and tricks like this, be sure to subscribe.
119
We'll see you in another video.

הורד אפליקציה

כל מה שאתה צריך כדי לדבר בשטף

הגיית AIציון לכל משפט
תרגול IPAשלוט בכל צליל
אוצר מיליםבנה את מאגר המילים שלך
משחק אוצר מיליםלמד תוך כדי משחק
FOR YOU

המלץ על סרטון

אוצר מילים והערות הגייה לשיעור הזה

שיעור הדיבור הזה מבוסס על הסרטון “Claude Code Is Wasting Tokens... Microsoft Built a Fix”. המילים שחוזרות הכי הרבה: token, agent, economizer, tool, repo. בסרטון הזה יש 119 משפטים ו-1344 מילים לתרגול Shadowing. הדיבור נמשך 7:29. הדובר מדבר מהר, כ-180 מילים בדקה, אז צפו לצלילים מחוברים ומקוצרים. 82% מהמילים נמצאות בין 3,000 המילים הנפוצות ביותר באנגלית; כדאי ללמוד את השאר לפני שמתחילים.

אוצר מילים מרכזי בסרטון הזה

15 מילים מהסרטון שכדאי ללמוד, עם ההגייה והמשמעות שלהן:

מילההגייהמשמעות
token שם עצם/ˈtoʊkən/אֲסִימוֹן, זִ׳יטוֹן
install פועל/ɪnˈstɔl/התקין
cloud שם עצם/ˈklaʊ̯d/עָנָן, עָב
skill שם עצם/skɪl/מְיֻמָּנוּת, כִּשּׁוּרִים
subscribe פועל/səbˈskɹaɪb/נִמְנָה
catalogue שם עצם/ˈkæt.əˌlɔɡ/קָטָלוֹג
folder שם עצם/ˈfoʊldɚ/תִּיקִיָּה
trick שם עצם/ˈtɹɪk/טריק, תכסיס
compare פועל/kəmˈpɛɚ/השווה \ הִשְׁוָה
solve פועל/sɒlv/פָּתַר
decline פועל/dɪˈklaɪn/ירד
bench שם עצם/ˈbɛnt͡ʃ/סַפְסָל
fewer/ˈfjuː.ɚ/פחות
default שם עצם/dɪˈfɔlt/ברירת מחדל
restore פועל/ɹɪˈstoɹ/הֶחְזִיר

Phrasal verbs שתשמעו

מילההגייהמשמעות
come out פועל/ˌkʌm ˈaʊt/יָצָא
turn off פועלכיבה, כִּבָּה

משפטים שכדאי לחזור עליהם

משפטים קצרים ושלמים מהסרטון שאפשר להשתמש בהם בשיחה יומיומית:

  • We're going to run our git clone.
  • Then I'll refresh the tool catalog.
  • That's why I say run it for a week.
  • If not, you've got the answer.
  • We'll see you in another video.

דקדוק בסרטון הזה

המבנים שהדובר משתמש בהם הכי הרבה, עם המילים המדויקות מהסרטון:

מבנהבסרטון
הווה מושלם (Present perfect) have/has + past participle — פעולה שקרתה בעבר ועדיין יש לה חשיבות או השפעה עכשיוhave been put · you've probably seen · has learned
סביל (Passive voice) be + past participle — הדגש הוא על מה שקורה, לא על מי שעושה את הפעולהhave been put · are turned · is called

הגייה ששווה לשים לב אליה

הדובר משתמש ב-35 קיצורים וצורות מקוצרות, כמו I'm, you're, I'll. אמרו אותם בצורה הקצרה, כפי שאתם שומעים אותם.

  • הצלילים של “sh” ו-“zh”: combinations /kɑmbɪˈneɪʃənz/, refresh /ɹɪˈfɹɛʃ/, extension /ɪkˈstɛnʃən/, intervention /ɪntɚˈvɛnʃən/, instruction /ɪnˈstɹʌkʃən/
  • מילים ארוכות — שימו לב להטעמה: combinations /kɑmbɪˈneɪʃənz/, intervention /ɪntɚˈvɛnʃən/, consistently /kənˈsɪstəntli/, eliminate /ɪˈlɪm.ɪ.neɪt/, repository /ɹɪˈpɒz.ɪ.t(ə.)ɹɪ/

צלילים שדוברי עברית מתקשים בהם:

  • /w/ — שפתיים מעוגלות, לא /v/: straightforward /ˌstɹeɪtˈfɔː(ɹ)wə(ɹ)d/, workflow /ˈwɝkfloʊ/
  • /æ/ — פתוח יותר מה-“אֶ” העברית, לא כמו “a” ב-“cup”: catalogue /ˈkæt.əˌlɔɡ/, variable [ˈvæɹ.i.ə.bl̩]
  • /r/ — הלשון לא נוגעת בשום מקום, לא ה-“ר” הגרונית: recover /ɹɪˈkʌvɚ/, micro /ˈmaɪ.kɹoʊ/, subscribe /səbˈskɹaɪb/, prompt /pɹɑmpt/, straightforward /ˌstɹeɪtˈfɔː(ɹ)wə(ɹ)d/

איך לתרגל עם הסרטון הזה

  1. הקשיבו לכל הסרטון פעם אחת בלי לדבר, ורשמו את המילים שאתם לא מכירים.
  2. התחילו במהירות 0.75×, תרגלו Shadowing משפט אחר משפט, וחזרו למהירות רגילה כשזה מרגיש קל.
  3. הקליטו את עצמכם והשוו למקור, תוך תשומת לב למילים כמו token, install, cloud.

מהי טכניקת השאדואינג?

שאדואינג היא טכניקת למידת שפה מבוססת מדע שפותחה במקור להכשרת מתורגמנים מקצועיים ופופולרית על ידי הפוליגלוט ד"ר אלכסנדר ארגואלס. השיטה פשוטה אך עוצמתית: אתה מקשיב לאודיו באנגלית ילידית וחוזר עליו בקול רם מיד — כמו צל העוקב אחרי הדובר עם עיכוב של 1–2 שניות בלבד. בניגוד להאזנה פסיבית או תרגילי דקדוק, שאדואינג מכריח את המוח והשרירים בפה לעבד ולשחזר דפוסי דיבור אמיתיים בו זמנית. מחקרים מראים שזה משפר משמעותית את דיוק ההגייה, האינטונציה, הקצב, הדיבור המחובר, הבנת השמיעה ושטף הדיבור — מה שהופך אותו לאחת השיטות היעילות ביותר להכנה לIELTS Speaking ולתקשורת באנגלית בעולם האמיתי.

טכניקת Shadowing: קרא את המדריך המלא שלב אחר שלב →