AI Engineer

Salary: $149,000 - 189,000 per year Requirements:

• We need at least 3 years of professional Python experience, including production work building and operating services.
• We need at least 1 year of hands-on production experience with LLMs, including prompt and context engineering, tool/function calling, structured outputs, and RAG.
• We look for working knowledge of LangChain, LangGraph, or a comparable framework such as AgentCore Strands, CrewAI, or Semantic Kernel.
• We need experience with LLM observability tooling such as Amazon CloudWatch, LangSmith, Langfuse, MLflow, or OpenTelemetry.
• We need experience designing evaluation frameworks such as MLflow, DeepEval, LLM-as-judge, or multi-turn regression.
• We expect fluency with Git, Docker, and modern API frameworks.
• We value clear written communication and sound judgment about when something is ready to ship.
• A bachelors degree is not required; practical experience from bootcamps, self-directed learning, career changes, or non-CS technical backgrounds also counts.
• Strongly preferred: hands-on development experience with Amazon Bedrock and/or AgentCore, including runtime, gateways, memory, policy, guardrails, observability, awscli, and evaluations.
• Strongly preferred: experience with Snowflake, Snowpark, or Snowflake Cortex.
• Strongly preferred: fluency in reading and writing SQL, plus understanding of semantic models.
• Strongly preferred: familiarity with multi-agent patterns such as supervisor/router, subagent/handoff, reflection, and human-in-the-loop.
• Strongly preferred: a thoughtful perspective on where agents should and should not act, with confidence pushing back when an agent is not the right solution.
• Strongly preferred: experience in Loyalty, MarTech, AdTech, or another data-rich B2B domain.

• We build agent harnesses in Python using LangChain and LangGraph, including tool calling, structured outputs, retries, streaming, and memory.
• We package agent harnesses for the AgentCore Runtime with the right context, tools, skills, and subagents for production workflows.
• We write the tools and skills our agents rely on, including API integrations, SQL queries against Snowflake, and Snowflake-backed knowledge retrieval with strong contracts and Pydantic validation.
• We develop evaluation harnesses such as golden datasets, LLM-as-judge checks, and regression suites using AgentCore Evaluations, and connect them to CI.
• We implement guardrails around tool execution, including auth scoping, input and output validation, PII and prompt-injection protections, and hallucination mitigation.
• We own what we ship by prototyping, deploying through Amazon AgentCore, monitoring traces, and fixing issues when they arise.
• We partner with data engineers on Snowflake-backed retrieval patterns, including Cortex Analyst and Cortex Search Services.
• We help refine our internal engineering patterns as the stack evolves.

• API
• CloudWatch
• Docker
• Git
• LLM
• Langfuse
• MLflow
• OpenTelemetry
• Python
• RAG
• SQL
• Snowflake
• AI
• Support
• JSON

More:

We are Kobie, a National Top Workplace and Top Remote Workplace headquartered in St. Petersburg, Florida, with a flexible remote-first environment that supports work-life balance. We offer Flexible Time Off, nine company-wide holidays, and a broad benefits package focused on growth, development, and personal well-being. We are a values-led organization where everyone is considered a leader, and we build market-leading loyalty solutions for some of the worlds most recognized brands. Our team is collaborative, innovative, and focused on helping enterprises create lasting emotional connections with their customers. We are committed to diversity, equity, and inclusion, and we provide an environment where teammates can bring their authentic selves to work each day.

last updated 33 week of 2026