Senior Engineer, AI/ML

A2Z; United States; On-site; USD 160,000 - 190,000; Full time

Benefits offered by this job

Medical benefits: dental, vision

401k matching

Unlimited PTO

Ownership of AI surface roadmap

Job summary

A2Z Sync is seeking a hands-on software engineer to build production agents in a code-first framework, integrating Strands Agents and LangGraph on AgentCore with a focus on safe, cost-aware workflows. You will work with AWS serverless components to orchestrate tools calling production APIs and ensure observability.

You will own end-to-end agent workflows, tool interfaces, evaluation gates, and guardrails, shipping production-ready agents that touch real dealer data and live transactions.

Qualifications

• 5+ years of software engineering experience, including hands-on production experience building LLM agents in a code-first framework.
• Business-conscious judgment: you start from the problem and its value, not the technology.
• Strong Python; ability to read and contribute to Java (Spring Boot) services.
• Hands-on experience with AWS serverless (Lambda, EventBridge, DynamoDB, Step Functions).
• Experience with prompt engineering, tool/function-calling design, and evaluating LLM output quality in production.
• An appetite for a heads-down building role: you measure your week in shipped code.

Responsibilities

• Build production agents in Strands/LangGraph on AgentCore: agent logic, tool orchestration, and multi-agent workflows.
• Create tool interfaces: framework-agnostic Python tools, MCP servers, and AgentCore Gateway targets.
• Develop evaluation harnesses: offline/online eval pipelines and LLM-as-judge CI/CD gates.
• Implement guardrails and safe failure paths with approval gates and rollback strategies.
• Manage cost and latency with per-agent cost tracking and routing across models.
• Establish production observability with tracing and dashboards.

Skills

Python

Java (Spring Boot)

LLM agents

Prompt engineering

AWS serverless

Code-first mindset

Business-context judgment

Tools

Strands Agents

LangGraph

AgentCore Runtime

MCP servers

AWS Bedrock

OpenTelemetry

CloudWatch

Job description

We operate a multi-tenant automotive SaaS platform serving thousands of dealer groups across the United States, running on a native AWS (NAWS) backend: Lambda, EventBridge, DynamoDB, S3, Step Functions. That platform works. Now we are making it think: production agentic AI systems built code-first and deployed on AWS Bedrock AgentCore, connected to real systems, making decisions and executing workflows against real dealer data and real money.

This is a hands-on builder role. Expect the large majority of your time in code. But we are not hiring a pair of hands. Every agent you ship makes decisions that touch dealer revenue, so we need an engineer who understands the business problem before writing the technical solution: someone who asks what a workflow is worth, weighs build cost against dealer impact, and knows when the right answer is a simpler tool, or no agent at all.

What you will own:

Building production agents in Strands/LangGraph on AgentCore: agent logic, tool orchestration, and multi-agent workflows deployed on AgentCore Runtime with Memory, Identity, and Gateway.

Tool interfaces: framework-agnostic Python tools, MCP servers, and AgentCore Gateway targets that let agents safely call production APIs.

Evaluation harnesses: offline/online eval pipelines and LLM-as-judge CI/CD gates that catch regressions before dealers do.

Guardrails and safe failure: approval gates and rollback paths that determine whether an agent fails safe or fails loud.

Cost and latency management: per-agent cost tracking and multi-model routing so each workflow earns more than it costs.

Production observability: AgentCore session tracing, OpenTelemetry, and CloudWatch dashboards for every agent you ship.

Tech environment: Strands Agents SDK on AWS Bedrock AgentCore (Runtime, Gateway, Memory, Identity, Observability); MCP servers; Bedrock Guardrails; Anthropic models (Haiku, Sonnet, Opus); NAWS stack (Lambda, EventBridge, DynamoDB, S3, Step Functions, ECS Fargate, Aurora, API Gateway, CDK); Python primary, Java (Spring Boot) secondary.

You work within the platform's established architectural patterns, and produce deliverables using agentic frameworks, AWS serverless core, MCP/tool design, and observability without someone looking over your shoulder. Prompt engineering and tool-use design are core engineering disciplines here, not an afterthought.

Scope and scale:

5,000+ destination dealer tenants, each with isolated databases and per-tenant configuration.

Agent workflows that touch production transaction data: real dealer money, not a sandbox.

Tens of thousands of API requests across REST, SOAP, and event-driven integration surfaces that your agents' tools will call into.

A growing portfolio of production agents, each with its own cost, latency, and evaluation budget.

Your first 6 months:

Month 1: Ramp on the platform architecture, existing agents, and integration surfaces, and on the dealer workflows behind them, so you know what the agents are for, not just how they run. Ship your first tool integration (MCP server or Gateway target) to production.

Months 2-3: Own end-to-end agent workflows, from design through production on AgentCore: Strands agent logic, tool interfaces, evaluation harness with CI gates, and guardrails.

Months 4+: Running autonomously: shipping new agents and tools with minimal oversight, with observability and cost tracking in place for everything you own.

Requirements

Must have:

• 5+ years of software engineering experience, including hands-on production experience building LLM agents in a code-first framework. Strands Agents and LangGraph are equally acceptable, as are Google ADK, OpenAI Agents SDK, and comparable frameworks. If you have built and operated real agents in any of them, you will pick up Strands quickly.
• Business-conscious judgment: you start from the problem and its value, not the technology, and you have walked away from a clever solution because a simpler one served the customer better.
• Strong Python; ability to read and contribute to Java (Spring Boot) services.
• Hands-on experience with AWS serverless (Lambda, EventBridge, DynamoDB, Step Functions).
• Experience with prompt engineering, tool/function-calling design, and evaluating LLM output quality in production.
• An appetite for a heads-down building role: you measure your week in shipped code.

Strongly preferred:

• Direct experience with Strands Agents and/or AgentCore Runtime (or migrating agent workloads from Lambda/Fargate to a managed agent runtime); LLM-as-judge, agent-to-agent orchestration, and agent governance.
• Direct production experience with MCP (Model Context Protocol) servers.
• Comfort owning production observability (OpenTelemetry, CloudWatch) for the systems you build.

Nice to have:

• Automotive, fintech, or multi-tenant marketplace platform experience.
• Familiarity with Bedrock Guardrails, LLM-as-judge evaluation, model cost/latency optimization, and multi-agent orchestration patterns.
• Experience with data pipelines (Glue/Athena) or ETL/data lake tooling.

Benefits

About A2Z Sync

A2Z Sync is a fast-paced and innovative automotive SaaS company seeking to make life better for our customers. We offer you a fun, casual, and collaborative culture, while fostering an environment where you work hard, see your results, and feel your impact. We are committed to our employees, and this starts with providing benefits that allow you to care for you and your family.

Mission

At A2Z Sync, we replace the friction of disconnected systems with the velocity of a single platform. We integrate digital insights with in-store operations to deliver transparent transactions that bring clarity to the car buyer and increased profitability to the dealer.

Our Values: We Are DRIVEN

• Dealership Obsessed: We measure our success by the dealer's wins and the trust of their buyers, not just our own code
• Relentless Ownership: No lone wolves, but no pass-backs either. We don't say "that's not my job."
• Invent with Purpose: We don't chase "shiny" tech. We replace guesswork with intelligence, building the "data backbone" that turns raw information into a competitive advantage
• Value Every Perspective: We are Better Together. We check egos at the door
• Evolve or Evaporate: Change is our constant. We stay ahead by learning faster than the competition
• Now Over Next: Perfection is the enemy of progress. We prefer action over endless analysis

Here’s how we are doing it:

• A2Z Sync offers comprehensive medical, dental, and vision benefits
• Employer provided STD/LTD and life insurance
• Matching 401k plan
• Unlimited paid time off, including 10 paid holidays
• Real ownership of a high-stakes AI surface - your roadmap, your architecture decisions, your metrics
• The expected salary range for this role is $160,000 to $190,000 annually, commensurate with experience and qualifications
• A2Z Sync is an Equal Opportunity Employer and does not discriminate against any employee or applicant for employment because of race, color, sex, age, national origin, religion, sexual orientation, gender identity, veteran status, basis of disability or any other federal, state or local protected class