Sr AI Software Developer

Honeywell•Phoenix, AZ

11h•Hybrid

About The Position

As a Sr AI Software Developer here at Honeywell, you will play a crucial role in the development of advanced AI solutions that drive business insights, enhance decision-making processes and empower AI solutions. Your expertise will help in critical data science development activities across all AI modalities (classic, Gen and agentic) and data types (structured and unstructured). You will report directly to our AI Director and you’ll work out of our Phoenix, AZ or Charlotte, NC location on a hybrid work schedule.

Requirements

5 or more years of experience in software development, ideally building backend, middleware, or distributed systems.
Strong proficiency in Python or Java/TypeScript, with hands-on experience building REST APIs, event-driven services, and microservices.
Experience working with AI/ML application patterns , including RAG, vector stores, prompt orchestration, and model/LLM integration.
Practical experience deploying applications on AWS using services such as Lambda, API Gateway, ECS/EKS, Step Functions, SQS/SNS, or DynamoDB.
Familiarity with Databricks (Delta Lake, SQL Warehouses, Model Serving, Vector Search) and integration patterns for data access and processing.
Experience building data or AI workflows that interact with Snowflake , including secure queries, role-based access, and performance optimization.
Experience working with Dataiku APIs or automation for data workflows, scoring endpoints, or integration into AI services.
Strong understanding of API design, authentication/authorization, secure coding practices, and system observability.
Ability to collaborate in an agile, cross-functional environment with platform engineering, MLOps, data engineering, and data science teams.

Nice To Haves

Experience building integrations for agentic AI systems , including tool registries, function-calling logic, multi-step planning support, memory stores, and safety guardrails.
Experience with vector databases and retrieval frameworks (Databricks Vector Search, OpenSearch, Pinecone, Milvus).
Knowledge of LLM/GenAI concepts (prompt engineering, orchestration, multi-turn conversation flows, caching strategies, re-ranking).
Familiarity with CI/CD practices for AI services and collaboration with MLOps (MLflow, evaluation pipelines, quality gates).
Experience building observability into AI services—tracing, metrics, logs, alerting, and model/agent behavior monitoring.
Strong problem solving skills with an eye toward performance, scalability, and maintainability.
Curiosity, adaptability, and a passion for applying emerging AI technologies responsibly and effectively.

Responsibilities

Design and develop AI application services and middleware that connect classic ML models, GenAI/LLM systems, and agentic AI components to enterprise applications and workflows.
Build production‑grade RAG (Retrieval-Augmented Generation) services , including chunking pipelines, embedding APIs, retrieval endpoints, caching, re-ranking, and content policy enforcement.
Develop agent tool adapters and integration layers enabling AI agents to safely perform actions (e.g., Snowflake queries, workflow triggers, system updates) using secure, controlled APIs.
Implement policy, safety, and guardrail middleware that enforces PII protection, content moderation, compliance rules, and safe function execution for agentic systems.
Create event‑driven and asynchronous services using AWS-native capabilities for agent orchestration , callbacks, monitoring, and workflow routing.
Build microservices and SDKs that enable scalable, low‑latency interactions between AI models, vector databases, and enterprise systems.
Collaborate with AI Architects, Platform Engineers, MLOps, Data Engineers, and Data Scientists to ensure systems are reliable, secure, observable, and aligned with best practices.
Implement robust testing frameworks for AI-driven services including regression tests, guardrail tests, prompt and agent behavior evaluations, and functional correctness checks.
Participate in code reviews, architectural discussions, and continuous improvement initiatives to enhance the performance and reliability of AI-powered applications.