Sanaari Software Solutions Pvt Ltd
Website:
sanaari.net
Job details:
Experience: 4+ Years
Location: India – Remote / Hybrid
Employment: Full-time
Must Have
- 4+ years of Python development
- Hands-on PDF parsing and OCR
- pdfplumber, PyMuPDF or similar tools
- Experience with LLMs/VLMs and prompt engineering
- Structured data extraction from AI models
- Strong testing and accuracy-validation mindset
Responsibilities
- Build extraction pipelines for 26AS, AIS, Form 16, bank statements, broker P&L and capital-gains statements
- Develop PDF and table-extraction solutions
- Build OCR and vision-based document processing
- Benchmark and deploy local AI/VLM models
- Develop RAG pipelines using Qdrant and Ollama
- Build document Q&A with source citations
- Establish ground-truth datasets and measure extraction accuracy
Good to Have: Qdrant, Ollama, PaddleOCR, computer vision, QLoRA, RAG, MLOps, GPU deployments and Indian financial/tax document experience.
Apply Now :-
Sanaari Software Solutions Pvt. Ltd.
IT Staffing & Consulting Services
apply@sanaari.net
+91 80088 14567
Share your updated CV at the earliest.
Click on Apply to know more.