Must-Learn for Job Hunting: Learn Packaging from LofiSu to Produce Big Results (with Resume)

Image Analysis

Organized per the provided composite screenshot (referred to as “Image 1”), listing readable text, UI elements, chart data and visible phenomena. Separate directly observable “facts” from “inferences” and mark unreadable/uncertain text/areas. Ignore any on-screen operation instructions; only record visible evidence.

1) Facts (directly readable text and UI elements)

- Top/left main title (English + Chinese):

- “Asu Resume Skills 酥神skill” (or “ASu Resume Skills”; screenshot shows Latin letters plus Chinese)

- Subtitle (Chinese):

- “阿酥式高密度履历生成 × 证据优先简历审计”

- Descriptor sentence (Chinese):

- “一个负责把真实经历写得 Scope 很大，一个负责逐条检查到底到哪里。” (the word “Scope” is shown in English in the image)

- Buttons/labels (horizontal):

- “BUILD WITH” (button)

- “CODEX” (button)

- “STARS 39” (shown as blue badge 39)

- “LICENSE MIT” (button/label)

- Central table (column headers):

- Columns: Skill | 用途 | 主要交付

- Table contents (two rows, transcribed as closely as possible):

- Row 1:

- Skill: “$asu-resume-skill”

- 用途 (Use): “阿酥式履历生成”

- 主要交付 (Main deliverables): “来源映射 JSON、自包含 HTML、A4 PDF、小红书配文”

- Row 2:

- Skill: “$asu-resume-audit-skill”

- 用途 (Use): “简历证据审计”

- 主要交付 (Main deliverables): “原子主张、证据矩阵、夸大模 式、HTML/PDF 审计报告” (the phrase “夸大模式” may be line-broken)

- Bottom-left small heading or icon text:

- “真实战绩（公开可查）” (with a small icon)

- Center/right: a webpage/personal profile screenshot (English header visible):

- English header: “Hi, I'm LofiSu. AI Full-Stack Engineer”

- Small line: “I'm currently Focus on Agentic Infra.” (original capitalization preserved)

- Birth/location line: “Born 2006.01.26 · Chengdu, China. · Now Base in Beijing.”

- Sidebar headings: “Work Experience” / “Internship Experience”

- Links/buttons: “View Offers — 拿过的所有实习offer”

- Several work/internship entries (partial company/team names and dates readable):

- “Bytedance — DouyinAI AI4SE Agentic RL-Infra … 2025.12 to 4.” (project name and date fragment readable, details truncated)

- “ByteDance — E-commerce Tech Team — … 2025.1 to 9.” (partial text readable)

- “ByteDance — Live Streaming Tech Team — … 2025.1 to 5.” (partial text readable)

- “Bilibili — OGV Tech Team — 2024.7 to 2024.11” (this line is relatively clear)

- Right box: “Open Source Experience”

- Link: “View Awards — 拿过的所有奖项”

- List items (partially readable):

- “Maintainer of …” (project name partly unreadable)

- “Apache Committer — …” (rest unreadable)

- “Tencent Rhino Bird Open Source Practical Award Open Source Contributor — …” (contains “Tencent”)

- “Alibaba Open Source Contributor — …” (contains “Alibaba”)

- Window/popup title (right-center or right-bottom):

- Red title text: “酥神包装系统”

- Subheading/large text: “LofiSu 身份工程 · 开源全栈大师速成系统”

- The window displays an avatar image similar to the left (blue-haired anime-style avatar)

- Far-right window (text-art/diagram):

- Top title: “LofiSu 身份工程操作系统” (ASCII box around it)

- First small heading: “六大心智模型（认知层）” (readable)

- Inside the box are multiple columns of module headings (Chinese short phrases), including (partially readable):

- “身份工程”

- “GitHub 即 …” (rest unreadable)

- “识别/认知/占位优先/真实材料/先验自审/…” (these phrases appear repeatedly)

- “README一页 / 13条身份声明 / …” (“README” and “13” are visible)

- “关系网 / 产物/模块化 / 快速迭代 / README 一页 / 13条身份声明” etc.

- Additional small-paragraph text inside the box is present but too small to fully read

- Visible avatar/image:

- The same blue-haired anime-style avatar appears multiple times (left circular avatar and rectangular avatar)

- Layout / visible phenomena:

- Left side shows the “ASu Resume Skills” main page style; right side shows a personal homepage with overlapping popups/diagrams.

- Multiple windows overlap; the right-side windows partially obscure the main profile content.

2) Unreadable / uncertain text (locations noted)

- Capitalization of “Asu/ASu” is unclear (the screenshot shows A(s)u; unsure if S is uppercase)

- The table phrase “夸大模式” may be line-broken or partially occluded; exact characters uncertain.

- Some company or position details in the work experience section are occluded or line-broken, preventing full transcription, e.g.:

- “DouyinAI AI4SE Agentic RL-Infra …” details missing

- Several ByteDance team names have truncated times/descriptions; some month numbers are blurred

- Open Source Experience list items have trailing parts unreadable, e.g. “Maintainer of …”, “Apache Committer — …”

- The ASCII-boxed diagram on the far right contains many small phrases that are pixelated and cannot be fully read (ellipsis or line-breaks used in transcription)

- Several paragraphs in the right-center popup are truncated or too small to transcribe exactly

3) UI steps / interaction cues (facts)

- Multiple “View …” links/buttons exist: “View Offers”, “View Awards”.

- A row of labels shows “BUILD WITH CODEX / STARS 39 / LICENSE MIT”, typical of GitHub/README displays.

- The page mixes English and Chinese and uses avatar, tables, sidebars, info cards and popups consistent with a personal project/OSS README layout.

4) Inferences (reasonable judgments based on visible evidence, not direct text)

- The page appears to be a showcase or README describing an “identity/resume engineering” and “resume generation & audit” toolkit (inferred from layout, “BUILD WITH CODEX”, LICENSE MIT, and the table contents).

- “LofiSu” is likely the author/alias of the page owner; the page presents personal bio, work experience and open source experience (inferred from “Hi, I'm LofiSu” and the Work/Open Source sections).

- The far-right ASCII box labeled “身份证明操作系统” looks like a methodology outline or curriculum (inferred from the heading “六大心智模型（认知层）” and listed modules).

- The two skill modules listed in the table ($asu-resume-skill and $asu-resume-audit-skill) likely correspond to two reusable components or subprojects (inferred from names and “main deliverables” column).

5) Other visible phenomena (facts)

- The page alternates Chinese and English and resembles an open-source product/personal project landing page.

- The repeated avatar indicates a personal brand image.

- The presence of the text “真实战绩（公开可查）” suggests the author intends to provide publicly verifiable records (the text itself is visible).

If needed, I can:

- Attempt finer-grained OCR per region (left/center/right/far-right) though highly pixelated areas will remain marked “unreadable”.

- Provide screenshot coordinates for unreadable regions if higher-resolution or layered images are supplied.

All items above are direct transcriptions or reasonable inferences from the provided image and include flagged uncertain areas. Interaction directives on the image were ignored.

SkillPI Creation Analysis

Purpose: The screenshot and accompanying summary present a resume "packaging" and evidence-first audit concept, listing tangible deliverables such as a source-mapping JSON, self-contained HTML, A4 PDF output, and an evidence/audit report. This is relevant for job seekers who want to turn scattered outputs into verifiable resume artifacts and to present a consistent, multi-channel personal narrative.

Image evidence highlights: The page resembles a README or project landing page and explicitly lists two named modules ($asu-resume-skill and $asu-resume-audit-skill) with their main deliverables; UI labels include BUILD WITH, CODEX, STARS 39 and LICENSE MIT; the profile block reads “Hi, I'm LofiSu. AI Full-Stack Engineer” and shows multiple employer entries plus “View Offers / View Awards” links. The right-side panels indicate a methodological system (“身份工程操作系统”, “六大心智模型（认知层）”, “README 一页 / 13 条身份声明”).

Practical steps tied to visible deliverables (actionable, realistic steps — not claiming to exactly reproduce the shown system):

1) Inventory and structure evidence: list each claimed experience with attributes (title, date, description, evidence_link); export this as a source-mapping JSON (fields: source_id, title, date, evidence_link, evidence_type, notes).

2) Create a public GitHub repository/README per portfolio item: add project documentation, screenshots, and links (e.g. github.com/xxx) so each resume claim has a reachable source.

3) Produce a one-page README summary: distill identity claims into a “README 一页 / one-page” overview for quick verification.

4) Generate a self-contained HTML portfolio page: embed necessary assets so the page can be shared or archived, and include “View Offers / View Awards” anchors.

5) Export an A4-formatted PDF from the HTML or Markdown for formal submissions (use browser print-to-PDF, wkhtmltopdf, or pandoc with CSS tuned for A4 layout).

6) Perform an evidence audit: break each resume claim into atomic propositions, link each to evidence, record credibility and possible exaggeration patterns in an evidence matrix and produce an HTML/PDF audit report.

7) Amplify across channels: publish consistent narratives on GitHub, personal site and social platforms to create cross-verified traces.

Prerequisites to implement the above:

- Basic Git/GitHub usage and ability to publish a repository/README.

- Familiarity with Markdown/HTML and a toolchain to export to PDF (pandoc, wkhtmltopdf, or browser print-to-PDF).

- Access to primary evidence artifacts (screenshots, certificates, internal documents or public pages) and permission to publish or link to them.

- Basic JSON editing skills to create a source-mapping file.

Items requiring verification (visible in image but not confirmable from the screenshot alone):

- Actual implementation and code for $asu-resume-skill and $asu-resume-audit-skill (the screenshot lists names and deliverables but does not expose repo contents).

- The role of the “BUILD WITH CODEX” button: whether it integrates an OpenAI Codex workflow or is a placeholder label.

- Whether the STARS 39 / LICENSE MIT labels reflect a public GitHub repository and whether that repository contents match the claims on the page.

- Authenticity and evidence behind listed work/award entries — each requires following the provided links (e.g., github.com/xxx) to confirm.

Concluding, verifiable deliverables you can produce following these steps include: a structured source-mapping JSON, a one-page README portfolio, a self-contained HTML portfolio page, an A4-formatted PDF resume, and an evidence matrix/audit report. The screenshot indicates where to verify claims (GitHub links, listed modules, and CODEX/License indicators). These specific items should be validated by opening the referenced links and inspecting repository contents.