Senior Backend Engineer Python & FastAPI (Document AI Systems)
Experience Required: 8+ Years
Location: Remote
Working Hours: Full-Time (40 Hours per Week)
Shift Timing: 5:30 PM 2:30 AM IST
Role Overview
- We are seeking a highly experienced Senior Backend Engineer with strong expertise in Python and FastAPI to design, build, and scale a document-processing backend capable of handling large, complex PDFs and AI-driven classification workflows.
- This role emphasizes system reliability, performance, and architecture. The ideal candidate will have experience building backend services that process 200+ page documents, integrate with Large Language Models (LLMs), and support robust exception handling, auditability, and cost-efficient execution.
- A critical aspect of this role involves designing and managing document taxonomies that enable classification, multi-page document assembly, confidence scoring, and downstream compliance workflows.
Required Skills & Experience
- 8+ years of hands-on experience with Python and FastAPI
- Proven experience in PDF processing using libraries such as PyPDF2, pdf2image, or similar tools
- Experience integrating LLM APIs for document analysis and classification (GPT-4 Vision API preferred)
- Hands-on experience with Google Cloud Platform (GCP), including:
- Cloud Storage
- Firestore
- Cloud Tasks
- Strong understanding of asynchronous processing patterns, including:
- Background workers
- Queues
- Retry mechanisms
- Solid systems design experience with focus on:
- Error handling
- Data lineage
- Audit trails
- Experience designing or working with document taxonomies, classification schemas, or structured labeling systems for AI-driven workflows
- Hands-on experience with OCR and document-processing systems
- Experience in financial services or compliance-driven environments
Familiarity With
- Taxonomy versioning
- Confidence thresholds
- Rules-based validation frameworks
- Experience supporting loan-processing workflows by classifying, assembling, and validating borrower documents such as:
- KYC documents
- Income proofs
- Bank statements
- Agreements
- Ability to implement taxonomy-driven rules, confidence scoring, and exception handling to enable faster and auditable credit decisions.
Key Responsibilities
- Design and develop backend services to reliably process large PDFs (200+ pages)
- Implement AI-powered document classification using defined taxonomies and confidence-scoring mechanisms
- Design, maintain, and evolve document taxonomies that drive classification, grouping, and validation logic
- Support multi-page document assembly using taxonomy-driven rules and page-level signals
- Build robust exception detection and recovery mechanisms, including:
- Review workflows
- Fallback paths
- Automated error handling
- Implement comprehensive audit trails and data lineage tracking, including taxonomy versioning and classification rationale
- Optimize system architecture and performance to manage infrastructure utilization and AI inference costs
- Collaborate with cross-functional teams to deliver scalable and compliant document-processing solutions.
What We're Looking For
- A backend engineer who thinks in terms of systems and workflows, not just APIs and endpoints
- Strong ownership mindset toward failure handling, edge cases, and long-running processes
- Ability to transform ambiguous document inputs into structured, explainable, and auditable outputs
- Comfortable working in document-heavy, AI-assisted backend environments
- Strong problem-solving skills with a focus on scalability, reliability, and maintainability
(ref:hirist.tech)
