Agentic AI Engineer
Hoffman Mechanical Solutions, Inc., Charlotte, NC, United States
You’ll design, build, and implement autonomous AI agents that reason, plan, and take action across our business — agents that turn messy real‑world goals into executable steps, plug into our systems, and deliver an industry-leading experience for our employee-owners and our customers.
Day to day, you will:
Architect agentic systems from the ground up
— perception, planning, memory, and tool‑use components that complete multi‑step workflows autonomously across APIs, internal services, and enterprise applications.
Turn LLM capability into business leverage
by integrating frameworks like LangChain, Semantic Kernel, AutoGen, CrewAI, and LangGraph with our data hub and operational systems.
Make agents trustworthy in production
— build the monitoring, observability, evaluation pipelines, logging, and safety controls that let us deploy with confidence instead of crossed fingers.
Sit at the table with leadership, sales, and operations , translate what they actually need into agentic workflows, and prove impact with KPIs that move the business — not vanity metrics.
Run experiments, gather feedback, and iterate fast
to keep pushing agent performance forward.
Document the craft
— architectures, design decisions, limits, and lessons — so the whole company compounds on what you build.
Who you’ll be doing it with
This is a team sport, and we play it with people we genuinely like working with.
You’ll be part of a small, tight‑knit crew working directly with HVAC and construction industry experts, and business leaders who are passionate about our customers, teammates and business. The kind of team that whiteboards together, creates together, celebrates wins together, and gives each other real feedback without ego. We share ideas openly, pressure‑test each other’s thinking, and cover for each other.
Our leadership is hands‑on and accessible — not a layer of management you escape to, but teammates you build with. Sales and operations aren’t on the other side of a wall; they’re in the room helping you figure out what to build and then using what you ship. Because everyone here is an employee‑owner, we all work for each other.
If you do your best work around smart, curious, low‑drama people who care as much about the mission as the code, you’ll fit right in.
What you bring
A strong software engineering foundation with real production chops in
Python
and backend or data systems.
Hands‑on generative AI / agentic AI experience, including at least 1 year building with LLM‑based agents, tool use, or multi‑agent systems.
Working fluency with one or more agentic/LLM stacks
LangChain, Semantic Kernel, AutoGen, CrewAI, LangGraph
and a major cloud AI platform
Azure OpenAI, Google Cloud, AWS , or equivalent.
Solid grasp of
prompt engineering, RAG, and context/memory management
for LLM applications.
Experience integrating AI with
REST/GraphQL APIs, databases, message buses, and event‑driven or microservices architectures .
Familiarity with
MLOps/AIOps
— CI/CD, monitoring, cost controls, and lifecycle management for AI workloads.
You’ve designed or operated autonomous agents in complex domains (operations, customer support, finance, healthcare, marketing automation, etc.).
Background in applied ML, reinforcement learning, or planning/decision‑making algorithms.
Real‑world knowledge of
AI safety, governance, and compliance
— data privacy, access control, human‑in‑the‑loop workflows.
Why this role is different
You're an owner from day one.
We’re 100% employee‑owned — the tools you build create value you directly share in.
Entrepreneurial, not corporate.
Fewer meetings, faster decisions, short distance between idea and production.
Real problems, real users.
Your agents will be used by your own colleagues and our customers — you’ll see the impact, hear the feedback, and produce the fix the same week.
Industry‑leading ambition.
We’re not chasing AI trends. We’re building the tools our industry will wish it had — alongside teammates who are as fired up about it as you are.
"As an Agentic AI Engineer here, you’ll own the end‑to‑end lifecycle of AI agents that turn high‑level goals into concrete action — connecting LLMs with our data, tools, and products to deliver measurable impact for the people who own this company and the customers who count on us."
This job description is intended to provide information regarding the essential functions and basic duties of this position. It is not intended to be construed as an exhaustive list of all functions, responsibilities, skills, and duties that could be assigned. Additional functions and requirements may be assigned by management as it deems appropriate. This document does not constitute a contract and is subject to change at the discretion of Hoffman & Hoffman, Inc.
#J-18808-Ljbffr
Day to day, you will:
Architect agentic systems from the ground up
— perception, planning, memory, and tool‑use components that complete multi‑step workflows autonomously across APIs, internal services, and enterprise applications.
Turn LLM capability into business leverage
by integrating frameworks like LangChain, Semantic Kernel, AutoGen, CrewAI, and LangGraph with our data hub and operational systems.
Make agents trustworthy in production
— build the monitoring, observability, evaluation pipelines, logging, and safety controls that let us deploy with confidence instead of crossed fingers.
Sit at the table with leadership, sales, and operations , translate what they actually need into agentic workflows, and prove impact with KPIs that move the business — not vanity metrics.
Run experiments, gather feedback, and iterate fast
to keep pushing agent performance forward.
Document the craft
— architectures, design decisions, limits, and lessons — so the whole company compounds on what you build.
Who you’ll be doing it with
This is a team sport, and we play it with people we genuinely like working with.
You’ll be part of a small, tight‑knit crew working directly with HVAC and construction industry experts, and business leaders who are passionate about our customers, teammates and business. The kind of team that whiteboards together, creates together, celebrates wins together, and gives each other real feedback without ego. We share ideas openly, pressure‑test each other’s thinking, and cover for each other.
Our leadership is hands‑on and accessible — not a layer of management you escape to, but teammates you build with. Sales and operations aren’t on the other side of a wall; they’re in the room helping you figure out what to build and then using what you ship. Because everyone here is an employee‑owner, we all work for each other.
If you do your best work around smart, curious, low‑drama people who care as much about the mission as the code, you’ll fit right in.
What you bring
A strong software engineering foundation with real production chops in
Python
and backend or data systems.
Hands‑on generative AI / agentic AI experience, including at least 1 year building with LLM‑based agents, tool use, or multi‑agent systems.
Working fluency with one or more agentic/LLM stacks
LangChain, Semantic Kernel, AutoGen, CrewAI, LangGraph
and a major cloud AI platform
Azure OpenAI, Google Cloud, AWS , or equivalent.
Solid grasp of
prompt engineering, RAG, and context/memory management
for LLM applications.
Experience integrating AI with
REST/GraphQL APIs, databases, message buses, and event‑driven or microservices architectures .
Familiarity with
MLOps/AIOps
— CI/CD, monitoring, cost controls, and lifecycle management for AI workloads.
You’ve designed or operated autonomous agents in complex domains (operations, customer support, finance, healthcare, marketing automation, etc.).
Background in applied ML, reinforcement learning, or planning/decision‑making algorithms.
Real‑world knowledge of
AI safety, governance, and compliance
— data privacy, access control, human‑in‑the‑loop workflows.
Why this role is different
You're an owner from day one.
We’re 100% employee‑owned — the tools you build create value you directly share in.
Entrepreneurial, not corporate.
Fewer meetings, faster decisions, short distance between idea and production.
Real problems, real users.
Your agents will be used by your own colleagues and our customers — you’ll see the impact, hear the feedback, and produce the fix the same week.
Industry‑leading ambition.
We’re not chasing AI trends. We’re building the tools our industry will wish it had — alongside teammates who are as fired up about it as you are.
"As an Agentic AI Engineer here, you’ll own the end‑to‑end lifecycle of AI agents that turn high‑level goals into concrete action — connecting LLMs with our data, tools, and products to deliver measurable impact for the people who own this company and the customers who count on us."
This job description is intended to provide information regarding the essential functions and basic duties of this position. It is not intended to be construed as an exhaustive list of all functions, responsibilities, skills, and duties that could be assigned. Additional functions and requirements may be assigned by management as it deems appropriate. This document does not constitute a contract and is subject to change at the discretion of Hoffman & Hoffman, Inc.
#J-18808-Ljbffr