Tech Stack
Amazon RedshiftAWSCloudMicroservicesNeo4jPython
About the role
- PlanHub is the leading pre-construction SaaS platform and marketplace helping general contractors, subcontractors, and suppliers connect and grow their businesses.
- We are seeking an experienced Sr AI Engineer with a strong AI development and software engineering background; role focuses on AI strategy, rapid prototypes, AIOps/MLOps, and AI solutions on AWS.
- What you’ll be doing: identify problems and solutions based on our Product and Technology strategy; build rapid prototypes and POCs; develop MVP and production-grade AI/ML solutions; develop agentic AI; architect for performance, cost and safety on AWS; work with cross-functional teams; trust-first AI development; data quality practices and responsible AI; assist in choosing third party solutions; work with data engineering to improve pipelines; CI/CD pipelines for AI/ML.
- You’ll use and improve CI/CD pipelines for AI and ML development, using containers and microservices and AWS infrastructure and services that accelerate development
- What you’ll need on day one: 5+ years of Python development experience and FastAPI or equivalent; 3+ years of CI/CD pipeline automation; 2+ years of RAG; 2+ years experience with multimodal LLMs; 1+ years agentic AI workflows; 1+ years guardrails; 1+ years AI QA testing; 5+ years experience AI/ML customer facing; 3+ experience with Jupyter Notebooks; 10+ years CS fundamentals; 5+ years AWS cloud; 5+ years Agile; Excellent communication
- What will make you stand out: Graph-based RAG and knowledge graph development using Neo4j; vertical agents; model fine-tuning; Amazon Redshift, SageMaker; OCR and computer vision for blueprints
- What's in it for you: Inc. 5000 recognition; flexible time-off; medical, dental, vision, life; 401(k); remote US; travel to West Palm Beach; etc
Requirements
- 5+ years of Python development experience and FastAPI or equivalent
- 3+ years of CI/CD pipeline automation using containers and microservices
- 2+ years of RAG experience with advanced concepts such as HYDE, Semantic search, chunking tuning, re-ranking (graph-embeddings RAG is a major plus), with an industry-grade framework such as LangChain, Haystack, AWS Bedrock knowledge base, and strong sense of which embeddings model to use and associated configurations
- 2+ years experience working with multimodal LLMs
- 1+ years of experience developing production-grade agentic AI workflows and solutions using one or more of: Bedrock Agents, CrewAI, LangChain, LangGraph, Autogen, Strand, Claude, OpenAI or equivalent, preferably on AWS
- 1+ years of experience working with and configuring guardrails for bias, content relevance and accuracy or using Natural Language Inference tools to accomplish the same, with security and privacy practices against model jailbreaking, prompt injection, sensitive data leakage and other such practices
- 1+ years experience helping institute AI QA testing practices leveraging responsible AI and accuracy/relevance to automate quality testing in the AI CI/CD pipeline, including unit testing and automation of traditional AI testing seeking accuracy of models, identification of model and data drift
- 5+ years of experience development AI/ML solutions that are customer facing as well as that support internal business processes
- 3+ experience with Jupyter Notebooks or equivalent using a data warehouse/data lake to derive insights and inform solution development and thought leadership
- 10+ years experience with computer science fundamentals including high performance algorithms and data structures
- 5+ years in a cloud environment leveraging caching, data flows and managed services on AWS or other cloud provider
- 5+ years experience working in an Agile development environment leveraging Jira, Confluence, GiHub or equivalent
- Excellent communication and interpersonal skills
What will make you stand out
- Graph-based RAG and knowledge graph development experience using Neo4j or equivalent
- Development experience with vertical agents
- Model Fine-Tuning and Prompt Tuning: Experience with fine-tuning LLMs and automating prompt tuning processes, using frameworks
- Experience working with Amazon Redshift, SageMaker
- Experience working with unstructured contents and OCR as well as computer vision systems to understand blueprints