Senior AI Engineer

Newfold Digital

19d

About The Position

Newfold Digital is seeking a hands-on Senior Applied AI Engineer to build production AI experiences across Network Solutions. The role combines software engineering fundamentals with practical AI problem solving, delivering systems for conversational agents, business and website creation, knowledge and FAQ agents, content experiences, domain discovery, and intelligent automation. The engineer will work with senior engineers, architects, and product partners to turn customer and product problems into reliable, secure, observable, and cost-efficient AI systems. This applied engineering role focuses on building, measuring, operating, and continuously improving real AI products. The engineer will implement AI services and workflows, integrate models and tools, develop retrieval-augmented generation systems, evaluate quality and reliability, debug production issues, and support assigned features across APIs, data, queues, model integration, observability, and production operations. The role also uses AI coding agents to accelerate engineering work while maintaining testing discipline, security, and code quality.

Requirements

At least 5 years of professional software engineering experience building production backend, distributed, or cloud-based systems.
At least 1 year of hands-on experience building Applied AI, LLM, RAG, NLP, or agent-based applications; production experience is strongly preferred.
Strong Python skills, including FastAPI, asynchronous programming, Pydantic, SQLAlchemy or SQLModel, and production API development.
Good backend and distributed-systems fundamentals, including REST APIs, concurrency, background processing, caching, reliability, and production debugging.
Hands-on experience with RabbitMQ, Kafka, Azure Service Bus, or equivalent queue and messaging architectures.
Hands-on experience integrating LLM APIs and building structured-output, function-calling, tool-calling, or agent-execution workflows.
Experience building or contributing to a RAG or knowledge-grounded system, including retrieval, embeddings, indexing, context construction, and evaluation.
Working knowledge of PostgreSQL and data modeling, with exposure to Redis, pgvector, vector databases, or hybrid-search technologies.
Demonstrated regular use of AI coding agents such as Cursor, Claude Code, Codex, or equivalent tools as part of software engineering work.
Ability to turn product goals into testable technical tasks, validate assumptions, implement solutions, and collaborate with senior engineers on architecture tradeoffs.
Ability to use experiments and evaluations to compare approaches using measurable outcomes.

Nice To Haves

Understanding of prompting, structured outputs, tool schemas, context management, retries, fallbacks, rate limits, caching, and streaming responses.
Understanding of retrieval quality, grounding, hallucination mitigation, prompt-injection risks, tool authorization, and practical agent guardrails.
Experience with Docker, CI/CD, automated testing, logging, observability, distributed tracing, OpenTelemetry, Langfuse, or similar tooling.
Experience with Semantic Kernel, Microsoft Agent Framework, LangGraph, PydanticAI, or equivalent agent frameworks.
Experience building or consuming Model Context Protocol tools and servers, or working with model gateways such as LiteLLM.
Experience with Azure AI Search, pgvector, Chroma, Kubernetes, Azure, OCI, or similar production AI platform technologies.
Experience contributing to customer-facing AI products, open-weight model inference, AI safety controls, automated evaluations, or high-scale backend platforms.
Curiosity, strong problem-solving ability, solid software engineering fundamentals, and the ability to learn quickly and turn ideas into reliable production features.

Responsibilities

Translate product requirements into practical solutions using LLMs, RAG, tool calling, agents, deterministic workflows, or traditional software as appropriate.
Build production AI services using Python, FastAPI, asynchronous workers, PostgreSQL, Redis, queues, model APIs, and external tools.
Implement reliable tool-calling agents and multi-step workflows that interact with internal APIs, MCP tools, business systems, and knowledge sources.
Build event-driven workflows using RabbitMQ, Kafka, Azure Service Bus, or equivalent platforms, including retries, dead-letter handling, idempotency, and failure recovery.
Build and improve RAG systems, including ingestion, chunking, embeddings, hybrid retrieval, reranking, metadata filtering, context construction, and citations.
Integrate models from OpenAI, Anthropic, Google, xAI, and open-weight ecosystems, and help select models based on quality, latency, reliability, and cost.
Build and run AI evaluations using curated datasets, regression tests, retrieval metrics, LLM-as-judge techniques, groundedness checks, and tool-execution evaluation.
Debug hallucinations, retrieval failures, incorrect tool usage, agent loops, latency issues, provider failures, and unexpected inference costs.
Use AI coding agents such as Cursor, Claude Code, OpenAI Codex, or equivalent tools to accelerate implementation, testing, debugging, refactoring, and documentation.
Break product goals into manageable, testable technical tasks, validate assumptions, and prototype alternatives when the best approach is unclear.
Choose simple, maintainable solutions and avoid using an agent or LLM when a deterministic approach is better suited.
Measure and improve quality, latency, reliability, safety, and cost over time.
Own implementation across APIs, workflows, data, queues, model integration, evaluations, observability, and production support for assigned features.
Use AI coding agents to increase engineering velocity while preserving testing discipline, security, maintainability, and code quality.