Senior AI Engineer

Blue Orange Digital•New York, NY

13h•Remote

About The Position

Blue Orange Digital is scaling its AI practice and needs Senior AI Engineers to deliver production agentic and AI systems to client engagements. The work is a mix of deep, single-client transformations and portfolio-wide programs. Examples of what you could be shipping in any given quarter include: End-to-end AI transformations for mid-market SaaS companies replacing manual, human-in-the-loop workflows with fully orchestrated agentic operations (document processing, review cycles, customer-facing copilots) AI readiness assessments and phased implementation roadmaps across private equity and growth-equity portfolios, where a single engagement may span five to fifteen portfolio companies at varying maturity levels Production RAG and retrieval systems for knowledge-heavy domains such as financial services, legal, compliance, and regulated public-sector workflows Agentic tooling and MCP-based integration layers that connect LLMs to client systems of record, internal APIs, and third-party SaaS Evals, observability, and guardrail frameworks that take client-built prototypes from notebook demos to load-tested, monitored production services Internal AI enablement engagements — helping client engineering orgs stand up their first production agent platform, define patterns, and upskill their teams You will work as a senior IC inside a delivery pod — the core unit of how BOD delivers AI work. A typical pod is three to five people: AI Architect — owns the platform and data foundations AI Transformation Consultant — drives strategy, roadmap, change management, and executive alignment Senior AI Engineer(s) (this role) — owns the agentic and ML implementation workstream Additional roles as needed to scale the build Pods operate as a cohesive unit that tackles cutting-edge AI strategy and implementation end-to-end — from discovery and architecture through shipped, measured production systems — across a rotating portfolio of interesting clients. You will own the AI engineering workstream on your pod, partner daily with the Architect and Consultant as peers, and report into the Practice Lead.

Requirements

5+ years building production data and ML systems in Python; 2+ years specifically on LLM-based or agentic systems
Hands-on experience with at least one major LLM orchestration framework (LangChain, LangGraph, Langflow, Databricks Agent Framework, or equivalent)
Production experience with Databricks (Unity Catalog, Delta Live Tables, MLflow) or comparable lakehouse platforms such as Snowflake with dbt
Deep knowledge of RAG architectures, vector databases, and embedding pipelines
Proven track record taking AI systems from prototype to production, including evals, monitoring, and on-call ownership
Comfortable working directly with client engineering teams as a peer and a coach

Nice To Haves

Databricks, AWS, or Azure certifications
Experience with MCP, tool-calling protocols, or agentic protocol design
Background in security-aware AI engineering (prompt injection, data leakage, access control)
Multimodal AI experience across text, document, and image
FinOps experience optimizing model and compute spend

Responsibilities

Build production-grade agentic systems on Databricks and other lakehouse platforms, including orchestration frameworks, task runners, and monitoring layers
Implement RAG pipelines, vector stores, and retrieval architectures that hold up under real-world load
Stand up evals, observability, and guardrails for client-deployed AI systems
Design and integrate MCP servers and tool-calling layers between LLMs and client systems
Lead the AI engineering workstream on a client pod, partnering with the Architect on platform decisions and the Consultant on roadmap
Coach client engineering teams on production AI patterns, including prompt management, model routing, and FinOps
Contribute to BOD’s internal Edge product suite, including reference architectures and the Blueprint scan engine