Senior AI Engineer //Fulltime

1. Types of AI Worked on -- AI solutions lying on top of historical systems (ERP, DW, etc) and leveraging them as part of prompt generation seem to be a common theme. Agentic structures seem to be a common thread in candidates interviewed so far. Some exposure to solutions leveraging RAG that pull from a variety of data types (unstructured text, semi-structured text, tabular data, relational databases, graph-based data, ontologies/taxonomies/controlled variables, etc.). Someone just coming from prompt solutions (talking to an LLM for helpdesk, for example) won t work., As a Senior AI Engineer, you will design and build the intelligence layer behind these systems-defining how they reason, retrieve information, and operate reliably in production environments. You will design LLM-based architectures (RAG, agent orchestration, and evaluation frameworks) and bring them into production as scalable, reliable systems-where correctness and reliability directly impact real-world outcomes. This work spans the full lifecycle of AI systems-from structuring data and defining retrieval strategies, to shaping how models reason and make decisions, to ensuring outputs are accurate, grounded, and production-ready., AI System Design & Architecture

o Design and implement LLM-based systems, including RAG pipelines, agent workflows, and orchestration layers

o Define how AI systems reason through complex, domain-specific scenarios, ensuring outputs are accurate, grounded, and reliable

o Design how AI systems interact with users, ensuring outputs are interpretable, actionable, and aligned with real-world workflows

o Architect retrieval and context strategies (chunking, ranking, embeddings) that ensure AI outputs are based on correct and relevant information

o Contribute to defining platform-level AI patterns and architecture that scale across multiple products and use cases

Data & Intelligence Layer

o Solve data integration and entity resolution challenges across fragmented and evolving data sources

o Design AI systems that operate effectively within real-world constraints, including latency, cost, and reliability requirements

Production & System Reliability

o Build and maintain evaluation frameworks to validate AI system performance (e.g., grounding, hallucination mitigation, response quality) before production use

o Troubleshoot issues across model behavior, data pipelines, and system integration in production environments

o Develop and maintain testing strategies for both traditional software components and AI system behavior

o Build and maintain production-grade AI systems, ensuring scalability, performance, and operational stability

Collaboration & Continuous Improvement

o Collaborate with product and business stakeholders to translate real-world problems into AI system designs and behaviors

o Continuously evaluate emerging AI technologies and incorporate them where they provide measurable improvements to system capability

o Mentor team members and contribute to best practices in AI system design, evaluation, and implementation

2. Career Arc -- They say they would consider someone with 5-6 years overall, but that candidate would need to have a rockstar vibe to them in that period of time (AI-focused). So far, the candidates they have interviewed have 8+ years of experience overall (some were 20+), with 2-3 years in AI solutions and 3-10 years in ML prior to that, ideally, in support of Data Scientist teams (automating their models). I suspect the candidate who lands this will be someone with 8 years, as their base budget seems low for a heavy senior in AI or someone in a smaller market (so the base for AI hasn't jumped as it has on the coasts). Startup background (ideally with AI) is highly valued.
3. Comms -- Communication level ideally at a consultative level (1st language English fee) as the role would be working with internal and client stakeholders.
4. AI Infra Ownership Nice to Have -- The AI application infrastructure exposure is nice to have (knowing how to take a pilot and scale it to a customer-facing product in the AI space or having done that). I would take someone who worked within a production AI Infrastructure (didn t build/maintain it)
5. Needs to have Product Vibe (not model R&D) -- Not model trainers or researchers. Need to be part of teams building out AI-based solutions that ship to customers (leveraging the LLM), o Tier 1 Core AI System Design (Non-Negotiable)

o AI System Design & Reasoning

Ability to design how AI systems reason, not just consume APIs

Retrieval-Augmented Generation (RAG) system design (chunking, retrieval, ranking, context assembly)

Agentic architectures and workflow orchestration

Orchestration patterns that scale AI systems from single workflows to platform-level capabilities

LLM system design (RAG, retrieval strategies, embeddings)

o AI Architectural Judgment & Evaluation Frameworks

Ability to make architectural tradeoffs (prompting vs fine-tuning, retrieval vs generation, latency vs accuracy)

Evaluation frameworks for LLM systems (grounding, hallucination detection, offline/online validation)

Strong Python

o Tier 2 Data Driven AI Systems

o Experience evolving from traditional ML modern GenAI systems

o Data modeling and data integration across multiple sources

o Complex SQL

o Experience designing data flows that support real-time or near-real-time AI decision-making

o Ability to structure and model data so AI systems produce accurate, context-aware, and grounded outputs

o Experience with ML frameworks (TensorFlow, PyTorch, etc.) (secondary now)

o Prompt engineering and response shaping strategies

o Tier 3 Production & Infrastructure

o Cloud platforms (Google Cloud Platform preferred)

o Containerization (Docker, Kubernetes)

o Workflow orchestration (Airflow, etc.)

o Performance optimization (latency, cost, scaling of inference systems)

o Monitoring and observability for AI systems (latency, cost, output quality)

o Tier 4 Supporting Engineering Skills

o React / Node

o General full-stack dev

o CI/CD, DevOps

NICE TO HAVE REQUIREMENTS:

o Experience applying AI to complex, real-world domains with ambiguous or incomplete data

o Experience working with proprietary or messy operational datasets (non-clean, non-labeled environments)

o Experience building domain-specific AI systems

o Experience with entity resolution/knowledge graphs

o Familiarity with vector databases

o Experience designing AI evaluation or testing frameworks

o Exposure to agent frameworks (LangChain, LlamaIndex, etc.)

o JavaScript/TypeScript (if already covered via React/Node, not critical separately)

o Data modeling/data warehousing concepts

o Distributed data systems

o Application architecture knowledge (beyond practical experience)

o Data integration concepts (theoretical vs hands-on)

Competitive pay commensurate with experience

Bonus (see above for details)

Medical, Dental, Vision, Life insurance, AD&D, Disability, FSA (healthcare/childcare), HSA, 401k