Backend AI Engineer

AI Engineer What You'll Do

• 1. Model Integration:
  Deploy and optimize LLMs, embeddings, and RAG pipelines into production applications.
• 2. AI Microservices:
  Build scalable Python-based services for knowledge extraction, question answering, and domain-specific reasoning.
• 3. Evaluation & Monitoring:
  Implement frameworks to measure precision, recall, latency, and drift; set up alerts for model degradation.
• 4. Prompt & Context Engineering:
  Design context windows, prompt templates, and fine-tuning strategies for insurance-specific tasks.
• 5. Data Pipelines:
  Collaborate with data scientists to create deterministic, auditable flows from raw documents to model-ready inputs.
• 6. Security & Compliance:
  Ensure models are deployed with guardrails, logging, and data isolation to meet SOC2/HIPAA standards.
• 7.
  
  Cross-Functional Collaboration:
  
  Partner with product, compliance, and engineering to ship features that balance innovation with reliability.

• 1.
  
  Experience:
  
  4+ years in applied AI/ML engineering, with production experience (not just prototypes).
• 2. Core
  
  Skills:
  
  Deep proficiency with Python, FastAPI/Flask, and ML libraries (Transformers, Lang Chain, PyTorch/Tensor Flow).
• 3. LLM Expertise:
  Hands-on experience integrating LLM APIs, fine-tuning, embeddings, or retrieval systems.
• 4. Cloud & Infra:
  Familiarity with deploying AI workloads on AWS/Azure/GCP; exposure to vector DBs (Pinecone, Weaviate, FAISS).
• 5. Evaluation Mindset:
  Experience building structured evaluation and monitoring for AI systems.
• 6. Startup DNA:
  Comfortable with ambiguity and rapid iteration; thrives in lean environments.

• 1. Background in insurance or financial services data.
• 2. Experience with orchestration frameworks (Lang Graph, Airflow, Prefect).
• 3. Familiarity with secure deployment patterns (VPC, single-tenant isolation, audit logging).