首先律師跟雇主和申請人討論一番，例如你的工作內容、職稱、未來加薪的幅度，為什麼會問到之後的薪水，因為這個階段的第一步是遞交你的工作職稱、內容、薪水是否能符合美國勞工部（U.S. Department of Labor, DOL）的標準（Prevailing Wage），也就是說你的工作和薪水不符合（薪水太低）就會導致審核不通過。同時薪水的標準會依地方不同而變化，（是不是覺得似曾相似，它類似 H1B 的第一步 LCA 申請）
以下管管道都可以 (1)job fair (2) on-campus recruiting (3) job search website other than the employer’s (4) employer’s website (5) employee referral program with incentives (6) private employment firm (7)trade or professional organization (8) campus placement office (9) local and ethnic newspaper (10) radio and television
*如果這個階段被裁員，那需要重新跑第一階段；如果 I-140 過了，雇主沒有去 withdraw I-140（很多公司不知道或是不在意）則可以找其他的公司繼續跑完後面流程，不需要重新跑 PERM。而I-140的180天後，即使原雇主 out of business 或是 withdraw 都不會使你的 I-140 失效。
由於兩種方式都各自有風險，H1B 的風險在於如果沒有抽中的不確定性，綠卡的風險在於等待時間長，如果超過 OPT 給的 3 年，就得回台灣等待再回美國。（This plan serves to mitigate risks associated with each path. The main risk of applying for H-1B is the uncertainty of success. The main risk of applying for GC is the long wait time (which may exceed my current Visa – 2.5 years).
11/12/2020 提出綠卡和 H1B，公司表示會聯絡律師 02/05/2021 開始比較有眉目（其中來來回回我就反覆問 I wanted to follow up and see if you have any updates around sponsorship） 02/18/2021 決定要辦綠卡，沒有工作簽證 H1B 03/2021 中間繼續來回，律師發資料給我和公司填寫 *同時請公司註冊成 E-verify company，為了符合 OPT extension 05/18/2021 In Process – PWD 02/01/2022 Need more information – PWD 不通過，要求公司給我能匹配職位的薪水 02/22/2022 Pending REDETERMINATION – 公司要求重新審核 05/09/2022 REDETERMINATION AFFIRMED – PWD 確定不通過
02/21/2022 In Process 重新遞交新的職稱和薪水給勞工局審核 08/23/2022 DETERMINATION ISSUED 09/08/2022 公司正式跟律師簽約下一步 09/28/2022 start to recruit 01/31/2023 finish recruitment 03/01/2023 summit final draft ETA-9089 04/30/2024 PERM 通過 06/17/2024 送出 i140 (premium) 06/26/2024 i140 Receive 07/03/2024 i140 Got Approve notice 07/18/2024 送出 i485 07/22/2024 通知信用卡沒刷過，重新寄出 08/05/2024 USCIS 收到並給通知和 case number 08/19/2024 收到指紋通知 Biometric Appointment 08/29/2024 打指紋 10/09/2024 Case Status: New Card Is Being Produced 10/11/2024 Case Status: Case Was Approved 10/15/2024 Case Status: Card Was Mailed To Me 10/21/2024 律師收到綠卡，確認住址寄出 10/22/2024 omggggg got my green card!!
我的心境
我覺得耐心極為重要，再來就是保持佛系心態，不然非常焦慮；一開始需要鼓起勇氣提出 H1B 和綠卡這件事，但是我拿這個offer 的前有先問過公司會不會 sponsor，因為我同時拿到另一個 offer 薪水比較高，但是確定不可能 sponsor，而我現在的公司我他完全不知道這是什麼，他仍然說好。到了 11 月多，我覺得應該要跟他們說，即使 H1B 是 3 月抽，可是先讓他們了解、內部討論以及知道價格看有沒有預算都需要時間。果不其然我提了後，無消無息好一陣子，就算我天天掛在心上也不好意思每天問，我固定兩個禮拜問一次，他們後來寄信給我在簡報上推薦的律師，其中一位我之前聯絡過，他也有傳訊息跟我說公司有詢問他 which made me feel better，後來 2 月公司決定跟這位我聯絡過的律師合作，然後他們想先幫我辦綠卡，因為他們認為這樣我才可以一直在美國工作，也省去 renew H1B 的麻煩🙏🙏🙏
學校的信箱可能被刪除，至少我們學校是（University of Texas at Dallas），如果有什麼重要的信，記得通知他們轉到其他的信箱，而 Amazon、PayPal、Venmo 等各種就學期間利用學校信箱申請的帳號都記得去更新，避免造成認證等困難。另外，我們學校提供的硬碟空間也會被關閉。
到學校 portal 上傳 USCIS approval notice
如果身分是從 STEM extension 到 H-1B，記得填寫 I-983 的第五頁 Final Evaluation，並一起上傳
Petition Letter 公司介紹 上一年度財務報告或年度報告（填寫表格用） copy of 公司章程（option） copy of 辦公室租賃（option） 如果公司之前沒有申請過 H1B，需要交 IRS 的稅號文件（EIN）
個人需要準備的文件：
最近一次的 copy of I-94 copy of Passport ID page showing the identity, expiration date, current US visas, most recent U.S. visa stamps 過往的 copy of I-20（持 F1 或 M1 簽證的學生適用） copy of I-983（Stem OPT Extension 學生適用） copy of IAP-66 或 SEVIS DS-2019（持 J1 簽證的訪問學者適用） copy of EAD 卡正反面 copy of I-797 最新版本的 resume（Word格式） copy of 學位證書（所有學位，包括學士學位、碩士學位和博士學位）和成績單 transcript 如果是在美國以外的國家獲得的學位，需提供學歷學位認證 Academic credential evaluation 曾經獲得的證書（如 CPA） job offer letter/招聘廣告和職位說明 job description 所有美國公司提供的 W-2 或 1099 最近三個月的工資單（需來自幫助辦理 H1B 的雇主） 原籍國的地址和郵遞區號（英文填寫） 住在美國的地址和郵遞區號
由於美國是醫藥分離的制度，所以醫院是不賣藥的，看完醫生後，護士通常會問 What pharmacy do you use? 接著你要到指定的藥局拿藥，像是CVS、Walmart 内都有藥局（醫生的收入跟開藥數量、價格沒有掛勾）此外如果你有預算考量，或是自己偏好的要藥品品牌，我們能在看診期間跟醫生提出，例如説希望不要選太貴的藥（相同的藥不同的牌子價格會不同，我試過！真的可以）。
In the wake of the 2024 election, the political map of the U.S. House of Representatives has never looked so fragile. With the Republican party holding a razor-thin majority of roughly 220 seats to 215, the battle for the 2026 midterms has, in effect, already begun. The most decisive engagements in this war, however, are not being fought at campaign rallies but in state courtrooms and closed-door legislative sessions.
The trend I am watching closely is the growing weaponization of partisan, mid-decade redistricting in the fight for House control.
Traditionally, congressional maps are drawn once after the decennial census (most recently in 2020) and remain fixed for the next ten years. But the near-tie of the 2022 election results has changed the game. With every single seat capable of deciding the majority, both parties are now aggressively litigating to force favorable maps into place before the 2026 election.
This “map war” is playing out across the country. In New York, Democrats are litigating to gain several seats. In North Carolina and Ohio, Republicans are working to solidify their own gerrymanders. Federal courts have forced redraws in Alabama and Louisiana. This battle is also creating internal party friction: in states Democrats control, like Maryland and Illinois, national leaders are pushing for aggressive new maps, while some state officials are resisting, fearing the high legal and political risks.
As seen in states like Texas and California, the party in control of state government often uses that power to draw lines that maximize its own advantage. The direct consequence of this “map game” is that it hollows out the meaning of the November general election.
This exacerbates a related trend: the transfer of political power to party primaries. When a district is gerrymandered to be “safe” (e.g., R+20 or D+20), the general election becomes a rubber stamp. The only election that truly matters is the primary. This explains why the debate over “open” versus “closed” primaries has become so intense; as gerrymandering makes the general election irrelevant, the rules of the primary become the only rules that matter.
Looking ahead to 2026, the outcomes of many elections may be decided by lawyers, judges, and map-makers long before the first vote is cast. This structural manipulation not only fuels political polarization—as candidates only need to appeal to their party’s most animated primary voters—but it also raises fundamental questions about whether a voter’s power is truly being counted.
Following President Trump’s 2025 proposals for high tariffs—including a potential 60% tariff on Chinese goods and a 10% universal baseline—many economists have focused on the high risks, such as inflation and trade wars. For a political analyst, however, this policy should be understood differently: it is one of the most effective political messaging tools in modern American politics.
The tariff strategy is not designed to win debates at the Peterson Institute for International Economics; it is designed to win votes in specific counties in Pennsylvania, Michigan, and Wisconsin.
This “economic nationalism” is a direct, unfiltered appeal to a very specific and electorally crucial demographic: blue-collar, non-college-educated, and union-household voters in the post-industrial “Rust Belt.” These are the voters who swung the 2016 election and remain the most contested bloc in American politics.
The political genius of the tariff argument lies in its simplicity. It bypasses complex economic theory and creates a powerful, populist narrative:
It defines a clear villain: “Globalist” elites and foreign competitors (especially China) who “stole” American manufacturing jobs.
It presents a clear hero: A protectionist leader (“I am a tariff man”) willing to fight back on behalf of the “forgotten” American worker.
It offers a clear solution: The tariff itself, a tangible “wall” to protect American industry.
Whether this narrative is economically sound is, politically, almost irrelevant. The message itself—”I am fighting for your job”—is what resonates in communities in Lackawanna County, PA, or Macomb County, MI, which have seen industries decline for decades.
This strategy also places the Democratic party in a difficult political bind. Opposing tariffs allows them to be painted as “pro-China” or part of the “globalist elite” who don’t care about American jobs. This forces them to walk a difficult tightrope, often trying to argue for “smarter” trade enforcement rather than rejecting protectionism outright—a message that lacks the simple, emotional clarity of a high tariff.
Ultimately, the high-tariff policy is a masterclass in political signaling. Its success is not measured in GDP or trade deficits, but in its ability to solidify a candidate’s bond with a critical voting bloc in the handful of states that decide the Electoral College. It is a policy that functions, first and foremost, as a ballot.
在國際政治與經濟交織的角力中，一場由美國掀起的關稅戰爭，正悄悄改變世界格局。這次，我們邀請到現任美國奧克拉荷馬中央大學（University of Central Oklahoma, UCO）經濟學教授 Glenn（Kuang-Chung Hsu），帶領我們從經濟學的視角，深入解析貿易赤字、國債危機與全球供應鏈重組的脈絡。這場訪談，我們試著理解表象之下的真相，也試圖釐清，面對這波波動，應該如何思考。
🎧 完整訪談也可在 Podcast 收聽： EP67｜川普的關稅政策錯了嗎？美國經濟學教授從學術角度聊貿易戰 ft. Glenn
這並不是美國第一次以強硬手段解決貿易問題。1985 年，美日貿易緊張之際，美國主導簽訂「廣場協議 Plaza Accord」，迫使日圓升值以降低美國對日貿易逆差。短期內，美國逆差確實收斂，但日本隨後陷入資產泡沫崩潰，經濟停滯長達十年，被稱為「失落的十年」。教授認為，川普的作法幾乎是歷史翻版，卻忽視了當年結局的教訓。
Glenn（Kuang-Chung Hsu） 為美國奧克拉荷馬中央大學（University of Central Oklahoma）經濟學教授，專長領域包括國際貿易、勞動經濟學與能源產業分析。他的研究著重探討全球經濟問題、企業佈局與政府政策間的互動，並著重外包、工資結構與貿易衝突對勞動市場與宏觀經濟的長期影響。Glenn 教授曾多次發表於《Reviews of International Economics, The journal of international trade and economic development, Empirical Economics 、《Applied Economics》、《Energy Economics》等國際期刊。畢業於台灣大學農業經濟碩士，並於德州農工大學（Texas A&M University）取得經濟學博士學位。
《聖經》記載，耶穌在受難週的週四晚上與門徒共進最後的晚餐（就是達文西名畫《最後的晚餐》的場景）。他當時已預知自己將被出賣，而那位出賣他的門徒就是猶大。13 號星期五的由來也跟《最後的晚餐》有關，猶大是第十三位入席的門徒，也因此「13」成了不吉利的數字。 隔天星期五，耶穌被釘上十字架，這天被稱為「受難日」也就是 Good Friday，而三天後星期日，耶穌復活了。
On its surface, Florida’s Amendment 3 is a straightforward ballot initiative to legalize recreational marijuana. The policy debate is clear, and so is the high legal threshold: it requires 60% support to pass. For an elections analyst, however, the real story is not the policy itself, but its potent, structural impact on the 2024 election.
The initiative’s primary political function is not necessarily legalization, but mobilization.
Florida has trended Republican in recent cycles, making it a difficult state for Democrats at the top of the ticket. The party’s challenge often lies in energizing its base and, crucially, turning out low-propensity voters. Amendment 3 is a powerful tool to solve this exact problem. Support for marijuana legalization polls exceptionally high among Gen Z, Millennials, and progressive-leaning Independents—demographic cohorts that are notoriously difficult to get to the polls.
This creates the critical “spillover effect.” A 22-year-old voter in Miami or Orlando, who might otherwise have skipped the election, may be highly motivated to vote “Yes” on Amendment 3. Once they have their ballot, the political calculus is whether they will also vote “down-ballot” for other Democratic candidates.
While this increased turnout might not be enough to flip the state for Kamala Harris in the presidential race, its true impact will likely be felt in Florida’s closely contested U.S. House races. In an environment where control of the House of Representatives is decided by just a handful of seats, the mobilization of even a few thousand new, left-leaning voters in a key district could be the decisive factor. This transforms the initiative from a policy question into a strategic political weapon.
The 60% threshold makes passage a significant challenge. Governor DeSantis and state Republicans are actively campaigning against it, meaning the measure could very well fail.
Analytically, however, this almost doesn’t matter. Even if Amendment 3 fails to pass (for example, receiving 58% of the vote), it may have already succeeded in its political mission. By being on the ballot, it forces Republicans to spend resources fighting it, and it successfully draws thousands of younger, progressive voters to the polls—voters who likely would have stayed home otherwise.
Ultimately, Amendment 3 is a powerful case study in modern political strategy, demonstrating how ballot initiatives can be used to bypass a legislature and, more importantly, reshape the electorate for a specific election. This is a tactic that will undoubtedly be refined and redeployed in future election cycles.
我在美國機場不只看一次這個很酷的無店員商店，起初以為只有 Amazon Go 才有，後來發現有些超商有跟 Amazon 合作這個 Just Walk Out 技術，大家可以看到最下面 Just Walk Out 的影片。
沒有收銀員結帳
簡單介紹一下這個技術，Amazon 為了提升購物便利性，開發了 Just Walk Out 的技術，首先應用於 Amazon Go 商店。這項技術讓顧客能夠在沒有收銀員結帳的情況下，輕鬆購物並離開。商店內部裝置了多個感應器和攝像頭，當時他們說 Amazon 結合了深度學習（Deep Learning）、機器學習（Machine Learning）和感知融合技術（Sensor Fusion）來追蹤顧客的行動及他們所拿的商品。但是這陣子的新聞報導這項技術實際上依賴了大量的人力來確保交易的準確性。預計有超過一千名在印度的工作人員負責觀察和記錄影片，以協助處理某些不能被程式語言辨識的交易。
合作案例 – Amazon Go、超商、星巴克等
Amazon 並未將這項技術局限於自家的 Amazon Go 商店，而是將其推廣至其他零售場所，包括 Whole Foods 和 Amazon Fresh，以及第三方零售商。在紐約，星巴克與 Amazon 合作開設了一家結合了星巴克 app 與 Just Walk Out 技術的概念店，提供了類似於 Amazon Go 的無人收銀體驗 （Engadget）。另外，Amazon 也與英國的大型超市連鎖店 Sainsbury’s 合作，在倫敦的 Holborn Circus 開設了無需結帳的超商。
被質疑的數據隱私
過去 Amazon 聲稱在提供 Just Walk Out 技術的同時，都有遵守數據保護法律，以保護消費者的隱私。而與第三方合作的店家，Amazon 雖然可以獲取一定的操作數據，但所有通過此技術收集的數據都受到合作協議和相關法律的限制。然而，當報導報出其實這些資料並不完全交由人工智慧來分析，而是由人來整理數據時，就引起了更多隱私方面的擔憂。
Amazon 出新招 Dash Carts AI 購物車
Amazon 正重新審視其零售技術戰略，Amazon 開始在一些 Amazon Fresh 商店中逐步淘汰 Just Walk Out 技術，並引入 Dash Carts AI 購物車。這個 Dash Cart依然能享受免排隊結帳，同時增加互動和更多的功能，像是提供即時收據查看，顧客可以通過購物車上的觸控螢幕看到和管理他們的購物清單，包括商品價格和當前總金額。觸控螢幕上有 QR code，登入 Amazon 帳戶，能及時查看當下的優惠，並導航找到商店內的特定商品。 此外，Just Walk Out 技術將繼續在 Amazon Go 商店、英國的小型 Amazon Fresh 商店以及一些第三方零售商中使用。總之，Just Walk Out技術的實際應用情況比最初宣傳的全自動化程度要低，而且在推廣過程中遇到了技術和隱私方面的挑戰。
