Junior AI intern

About the Role

Secondz is seeking a Junior AI Intern to build AI-powered applications using Python, OpenAI APIs, document extraction, computer vision, and OCR. The role involves extracting and processing information from PDFs, Excel files, and images using AI models and document-processing tools.

Key Responsibilities

• Integrate OpenAI APIs into Python applications using API keys
• Build document extraction workflows for PDFs and Excel files
• Utilize vision-capable AI models to understand images, tables, and document screenshots
• Implement OCR workflows to extract text from scanned documents and images
• Process and validate structured data returned by AI models
• Convert extracted information into structured JSON formats
• Handle missing, inconsistent, or incorrectly extracted information
• Test extraction accuracy across different document formats and layouts
• Debug issues and improve the reliability of AI-powered workflows
• Write clean, reusable Python code
• Collaborate with the backend team to integrate AI features into applications

What You'll Learn

• Integrating AI models into Python applications
• Extracting data from PDFs, spreadsheets, and images
• Working with OCR and vision-based document understanding
• Designing reliable prompts and structured AI outputs
• Testing extraction accuracy and handling AI errors
• Building practical AI features that integrate with backend systems

Requirements & Tech Stack

• Basic to intermediate knowledge of Python
• Familiarity with OpenAI APIs and API-key-based integration
• Understanding of JSON and structured data
• Basic knowledge of PDF and Excel file processing
• Familiarity with libraries such as pandas and openpyxl
• Understanding of OCR and text extraction from images
• Interest in computer vision and vision-capable AI models
• Basic understanding of REST APIs, HTTP requests, and error handling
• Familiarity with Git and GitHub
• Willingness to learn, test, debug, and improve AI solutions

Good to Have

• Experience with OpenAI vision models or other multimodal LLMs
• Familiarity with OCR tools such as Tesseract, EasyOCR, or PaddleOCR
• Experience with PDF libraries such as PyMuPDF or pdfplumber
• Knowledge of FastAPI and PostgreSQL
• Familiarity with prompt engineering and structured outputs
• Basic knowledge of image preprocessing and table extraction
• Personal projects involving AI, OCR, document processing, or computer vision