Uploading Documents - Best Practices¶
Learn how to prepare and upload documents for optimal timeline extraction.
Supported File Formats¶
| Format | Support Level | Notes |
|---|---|---|
| ✅ Excellent | Text-based PDFs work best | |
| TXT | ✅ Excellent | Plain text with structure |
| DOCX | ✅ Good | Microsoft Word documents |
| MD | ✅ Good | Markdown files |
| ODT | ⚠️ Limited | May need conversion |
| Images | ❌ Not supported | Use OCR tool first |
Document Type Guidelines¶
Resumes (Best Results)¶
Why resumes work well:
- Clear chronological structure
- Explicit dates and durations
- Standardized sections
Optimization tips:
✅ DO:
- Use standard section headers (Experience, Education)
- Include month/year dates (Jan 2020 - Mar 2023)
- Separate entries with clear spacing
- Include locations (City, State/Country)
❌ DON'T:
- Use vague dates ("recently", "a while ago")
- Merge multiple jobs in one paragraph
- Skip locations
- Use images instead of text
Example format:
PROFESSIONAL EXPERIENCE
Senior Software Engineer
TechCorp Inc., San Francisco, CA
June 2020 - Present
• Led development team of 5 engineers
• Built ML pipeline processing 1M+ events/day
Cover Letters (Good Results)¶
Why they work:
- Narrative with timeline
- Specific achievements
- Clear transitions
Tips: - Mention specific dates when discussing experiences - Use paragraphs with clear timeframes - Reference positions and companies explicitly
Personal Statements (Good Results)¶
Academic/professional narratives:
- Research timelines
- Publication dates
- Conference presentations
Tips: - Include dates for all milestones - Mention specific projects/publications - Reference institutions and locations
Pre-Upload Checklist¶
Before uploading, verify:
- Dates are explicit - Month/Year minimum
- Text is selectable - Not a scanned image
- Structure is clear - Headings and sections
- File size < 10MB - For optimal processing
- Format is supported - PDF, TXT, DOCX, MD
Upload Process¶
Step 1: Access Upload Tab¶
Navigate to 📤 Upload & Process tab
Step 2: Choose Upload Method¶
- Drag file from file browser
- Drop onto upload area
- See filename confirmation
- Click "Browse files"
- Navigate to document
- Select and click "Open"
- Select multiple files (Ctrl/Cmd + Click)
- Upload in one batch
- Process sequentially
Step 3: Configure Settings (Optional)¶
Advanced Settings
AI Model: - gpt-3.5-turbo - Fast, cost-effective - gpt-4 - Higher accuracy
Temperature: - 0.1-0.3 - Deterministic (recommended) - 0.4-0.7 - Balanced - 0.8-1.0 - Creative (not recommended)
Max Tokens: - 1000 - Short documents - 2000 - Standard (default) - 4000 - Long documents
Step 4: Process¶
- Click 🚀 Process Documents
- Monitor progress bar
- Check success message
Optimizing Document Content¶
Date Formatting¶
Best formats:
✅ Excellent:
- January 2020 - March 2023
- 01/2020 - 03/2023
- 2020-01 to 2023-03
- Jan 2020 - Present
⚠️ Acceptable:
- 2020 - 2023 (year only)
- Q1 2020 - Q4 2023
- Spring 2020 - Summer 2023
❌ Poor:
- "recently"
- "a few years ago"
- "during college"
- "around 2020"
Event Descriptions¶
Effective descriptions:
✅ Good:
"Led team of 5 developers building cloud-native microservices
architecture serving 1M+ daily users across AWS infrastructure"
❌ Poor:
"Worked on some projects with the team"
Key elements:
- Action verbs (Led, Built, Managed)
- Quantifiable metrics (5 developers, 1M users)
- Technologies/tools (AWS, microservices)
- Impact/results (serving 1M+ users)
Location Information¶
Include when relevant:
✅ Complete:
Google, Mountain View, CA
Stanford University, Stanford, CA, USA
Remote (based in Seattle, WA)
⚠️ Partial:
Google (city missing)
Mountain View (company missing)
❌ Vague:
"Bay Area"
"US West Coast"
Handling Different Document Types¶
Multi-Page Resumes¶
Processing tips:
- Ensure all pages are included
- Keep consistent formatting across pages
- Use page breaks appropriately
- Verify file size < 10MB
Scanned Documents¶
If you must use scanned PDFs:
-
Use OCR first:
-
Verify text extraction:
- Open PDF in viewer
- Try selecting text
-
Ensure it's readable
-
Clean up OCR artifacts:
- Fix garbled characters
- Correct date formatting
- Review spacing issues
Multiple Language Documents¶
Current support:
- ✅ English (full support)
- ⚠️ Spanish, French, German (partial)
- ❌ Non-Latin scripts (limited)
Tips for multilingual:
- Keep dates in ISO format (YYYY-MM-DD)
- Use English month names when possible
- Translate key sections if accuracy critical
Batch Processing¶
Uploading Multiple Documents¶
When to batch process:
- Multiple resumes from different periods
- Related documents (resume + cover letters)
- Complete career documentation set
Best practices:
- Start with most important document
- Process in chronological order (oldest to newest)
- Review after each document before uploading next
- Use clear filenames (resume_2020.pdf, resume_2023.pdf)
Managing Duplicates¶
ChronoScope automatically detects similar events:
⚠️ Potential duplicate detected:
Event 1: "Senior Developer at TechCorp" (2020-2023)
Event 2: "Senior Software Engineer, TechCorp" (2020-2023)
Similarity: 87%
[ Keep Both ] [ Merge ] [ Delete Event 2 ]
Troubleshooting Upload Issues¶
File Won't Upload¶
Upload fails or hangs
Check:
- File size < 10MB
- Supported format (PDF, TXT, DOCX)
- Not a protected/encrypted PDF
- Internet connection active
Solutions:
- Compress large files
- Convert to supported format
- Remove PDF password protection
- Try different browser
No Events Extracted¶
Processing completes but no events found
Possible causes:
- Scanned image PDF (not text)
- Unclear date formatting
- Unstructured content
- Wrong document type
Diagnosis:
- Check if text is selectable in PDF
- Review date formats in document
- Try demo data to verify app works
- Check Advanced Settings for errors
Poor Extraction Quality¶
Many events missed or incorrect
Improve quality:
- Reformat document:
- Add clear section headers
- Explicit dates (Month Year)
-
Structured entries
-
Try different AI model:
- Switch to GPT-4 (higher accuracy)
-
Adjust temperature lower (0.1-0.2)
-
Manual intervention:
- Add missing events manually
- Edit extracted events
- Provide clearer source document
File Organization Tips¶
Recommended Structure¶
my-timeline-docs/
├── resumes/
│ ├── 2020_chronological_resume.pdf
│ ├── 2023_current_resume.pdf
│ └── academic_cv_2022.pdf
├── cover-letters/
│ ├── techcorp_application_2020.pdf
│ └── startup_cover_letter_2023.pdf
└── other/
├── linkedin_profile_export.txt
└── personal_statement_grad_school.pdf
Naming Conventions¶
Good filenames:
Poor filenames:
Next Steps¶
After uploading:
Quick Reference¶
| Task | Action |
|---|---|
| Upload single file | Drag & drop or Browse files |
| Upload multiple files | Ctrl/Cmd + Select multiple |
| Process documents | Click "🚀 Process Documents" |
| Check extraction quality | Review confidence scores |
| Handle duplicates | Use similarity detection |
| Improve results | Reformat document, try GPT-4 |
Pro Tip
Keep a "clean" version of important documents optimized for ChronoScope:
- Clear date formats
- Structured sections
- Explicit locations
- Quantifiable achievements