Senior AI Engineer

This is a fully remote position, open to applicants in Mexico.

📋 Description

• Create, develop, and sustain production-level LLM integration pipelines, which encompass retrieval-augmented generation (RAG), prompt engineering, output parsing, and chain orchestration.

• Design and manage AI functionalities within Jeeves's primary financial offerings, including spend categorization, document extraction, anomaly detection, financial Q&A, and automated reconciliation.

• Carry out structured output validation, fallback mechanisms, and confidence scoring to guarantee that AI decisions adhere to reliability standards for financial applications.

• Assess and incorporate AI frameworks and tools (LangChain, LlamaIndex, OpenAI API, Anthropic API, HuggingFace, vector databases) while advocating for the most suitable tools for specific tasks.

• Develop prompt versioning and evaluation methodologies to maintain the accuracy and consistency of AI outputs as models and data evolve.

• Design and sustain vector search pipelines utilizing databases such as Pinecone, Weaviate, or pgvector to enable semantic search and RAG-based functionalities.

• Construct document ingestion and chunking pipelines for Jeeves's financial data, processing invoices, receipts, policy documents, and transaction records.

• Enhance retrieval quality by optimizing embedding model selection, chunk strategies, metadata filtering, and re-ranking techniques.

• Collaborate with data scientists to transition trained ML models from experimental notebooks to production-serving infrastructure.

• Develop and maintain model serving endpoints that comply with appropriate latency service level objectives (SLOs), ensuring input validation and output monitoring.

• Execute model performance oversight and data drift detection to confirm that production models retain accuracy over time.

• Assist in model retraining processes by designing clean data pipelines and feature engineering that can be updated continuously.

• Seamlessly integrate AI services with Jeeves's backend microservices by establishing clear API contracts, circuit breakers, and graceful degradation patterns.

• Produce high-quality, testable backend code in Python or Go/Node.js to support AI-integrated functionalities.

• Equip AI components with structured logging, distributed tracing, latency dashboards, and alerting to ensure operational transparency.

• Develop human-in-the-loop review processes for AI decisions requiring oversight, particularly for significant financial actions.

• Collaborate with Product, Backend Engineering, and Data Science teams to outline the AI roadmap and convert requirements into dependable systems.

⛳️ Requirements

• A Bachelor's degree in Computer Science, Engineering, or a related discipline, or equivalent practical experience.

• Over 5 years of professional software engineering experience, with a minimum of 3 years dedicated to AI/ML systems in a production setting.

• Practical experience in building and deploying LLM-powered applications using APIs such as OpenAI, Anthropic, or Cohere in a production context.

• Proficiency in designing and managing RAG pipelines, including chunking strategies, embedding models, and vector database integration (Pinecone, Weaviate, pgvector, or similar).

• Strong expertise in Python for AI/ML tasks; familiarity with at least one AI orchestration framework (LangChain, LlamaIndex, or similar).

• Experience with ML model serving infrastructure, including REST or gRPC inference endpoints, input/output validation, latency budgeting, and monitoring.

• Solid understanding of backend engineering principles: REST APIs, relational databases (PostgreSQL preferred), asynchronous patterns, and cloud infrastructure (AWS, GCP, or Azure).

• Familiarity with observability tools: structured logging, distributed tracing, and building dashboards for monitoring AI system health.

• Background in fintech, financial services, or any regulated industry where AI reliability and auditability are paramount.

🏝️ Benefits

• Options for remote work.