Senior AI Engineer

This is a remote position.

Role Summary

We are hiring a highly skilled Senior AI Engineer to design, build, and scale enterprise-grade document intelligence systems powered by self-hosted LLM infrastructure. This is a hands-on individual contributor role focused on architecting and operating production-ready OCR-driven pipelines that process long-form, document-heavy workloads (200+ pages) with deterministic, auditable outputs.

You will partner closely with AI leadership while independently owning architecture decisions, system optimization, and production reliability across document processing and LLM inference systems.

Key Responsibilities

1. Enterprise Document Intelligence Architecture

• Architect and implement end-to-end pipelines for OCR-heavy, long-form document processing.
  
• Lead PDF ingestion, layout-aware parsing, and multi-page document reconstruction.
  
• Design chunking, segmentation, and metadata tracking strategies for large-scale workflows.
  
• Implement deterministic retry mechanisms, exception handling, and failure recovery systems.
  
• Optimize processing frameworks to handle 200+ page documents at scale with high reliability.
  
• Ensure system scalability, performance consistency, and operational stability under heavy document loads.
  

2. OCR & Structured Data Extraction Engineering

• Integrate and optimize OCR engines such as Tesseract, PaddleOCR, and layout-aware or vision-language models.
  
• Build layout-sensitive extraction systems leveraging bounding boxes and structural metadata.
  
• Implement deterministic schema validation, cross-field consistency checks, and rule-based validation layers.
  
• Minimize manual QA dependency through automated validation and control frameworks.
  
• Maintain traceability from every extracted field to its original document span for auditability.
  

3. Self-Hosted LLM Infrastructure (Production Ownership)

• Deploy and manage open-source LLMs using vLLM and Hugging Face TGI.
  
• Operate GPU-backed serving infrastructure for large-scale inference workloads.
  
• Optimize inference performance via KV cache strategies, intelligent batching, context window tuning, and throughput-latency balancing.
  
• Monitor GPU utilization and improve cost efficiency per inference request.
  
• Own uptime, observability, and reliability of LLM production environments.
  

4. Deterministic Validation & Governance Systems

• Design validation layers external to LLM outputs.
  
• Build schema enforcement engines, invariant checks, rule systems, and rejection logic.
  
• Implement automated exception routing without default human review.
  
• Ensure reproducibility, audit trails, and measurable correctness guarantees.
  
• Establish reliability standards for high-stakes and compliance-driven use cases.
  

5. Production Engineering & Scalability

• Architect systems capable of managing large document volumes and high concurrency.
  
• Handle complex failure states with robust fallback and recovery mechanisms.
  
• Build observability frameworks including structured logging, distributed tracing, and metrics monitoring.
  
• Improve monitoring, alerting, and operational tooling across AI document pipelines.
  
• Collaborate cross-functionally to deliver production-grade AI solutions.
  

Required Skills & Experience

• 6+ years of strong Python-based backend engineering experience.
  
• Proven expertise in building OCR-driven document intelligence systems in production.
  
• Experience handling long-form PDFs (100+ pages).
  
• Hands-on experience with:
  
• vLLM or Hugging Face TGI
  
• GPU-backed LLM serving infrastructure
  
• Open-source LLMs such as LLaMA, Qwen, Mistral, etc.
  
• Experience designing deterministic validation systems (schema + rule enforcement).
  
• Strong debugging, systems-thinking, and architectural decision-making skills.
  
• Ability to clearly communicate technical trade-offs and business impact.
  

Strongly Preferred

• Experience with layout-aware models such as LayoutLM, DocFormer, or vision-language architectures.
  
• Demonstrated experience optimizing GPU utilization and inference cost efficiency.
  
• Background in regulated industries (finance, healthcare, compliance).
  
• Familiarity with document-intensive workflows like underwriting, loan processing, or insurance claims.