Agentic AI Platform Engineer

Company Overview

Spectrix Analytical Services is seeking an Agentic Platform Engineer. This role focuses on building and operating the AWS platform that hosts AI agents, MCP servers, and agent tooling for discovery research groups. This is a full-time (40 hours/week) remote position open to candidates in the United States.

Working hours within one to two hours of US Eastern time are preferred; West Coast candidates who can overlap with Eastern business hours will also be considered.

The contractor will own the runtime, access control, and deployment path that agents run on. Agents and MCP servers built by the team call internal APIs, Snowflake, and scientific applications on behalf of individual scientists, so the platform has to authenticate users through Entra ID, scope what each agent can reach, keep credentials out of agent code, and record what each agent did.

In addition to the platform, the contractor will build MCP servers, agent skills, and other integrations as needed, and we expect the work to move between platform and integration as priorities shift.

Areas of work:

• Agent runtime. Stand up and operate the runtime that hosts agents and MCP servers, using Amazon Bedrock AgentCore or a comparable agent runtime, together with serverless (Lambda) and container (ECS or EKS) workloads. Define how a new agent or MCP server is packaged, deployed, versioned, and retired.
• Infrastructure as code. Write and maintain the Terraform modules for the runtime, networking, gateways, and supporting services, so that a new environment, AWS account, or region is created from configuration.
• Gateway and access control. Run the API gateway layer in front of agents and tools. Integrate Entra ID for user authentication, map Entra groups to the agents and tools each user can call, and implement token exchange so an agent calls downstream services with the requesting user’s permissions.
• Identity and secrets. Design IAM roles and workload identity for agents and MCP servers, manage credentials for Snowflake and internal APIs through AWS Secrets Manager or equivalent, and automate rotation.
• CI/CD and observability. Build the pipelines that test and deploy agents, MCP servers, and platform components. Instrument agent and tool calls with tracing, logs, and metrics so that latency, error rate, model usage, and cost can be reported per agent and per user group.
• MCP servers and integrations. Build MCP servers and agent skills that connect agents to Snowflake, internal APIs, and scientific applications. Provide templates, a local development path, and documentation so that other teams can ship their own MCP servers and skills onto the platform without platform changes for each one.

Key Responsibilities:

• Develop and maintain production Python services, MCP servers, platform tooling, and CLI utilities.
• Write and review Terraform for all platform infrastructure as part of daily workflow.
• Deploy and operate agent, serverless, and container workloads across AWS accounts and environments.
• Maintain the access-control model for agents and tools and review new integrations for how they handle identity, secrets, and data scope.
• Work directly with scientists, application developers, and IT security to decide what to build, and support adoption through documentation and hands-on help.

Qualifications:

• Bachelor’s degree with 7–10 years of relevant experience, or a Master’s degree with 5 or more years of relevant experience.
• Experience having built and run a shared internal platform used by other engineering teams, including its deployment path, support, and upgrades.
• Strong AWS experience with both serverless (Lambda, API Gateway, Step Functions or similar) and container (ECS or EKS) workloads.
• Experience with Amazon Bedrock AgentCore or a comparable agent runtime is strongly preferred.
• Strong Terraform experience, including module design and multi-account deployments.
• Experience with API gateways and access control using Entra ID (Azure AD), including OAuth 2.0 and OIDC flows.
• Experience with cloud identity and secrets management: IAM, workload identity, Secrets Manager or Vault, and credential rotation.
• Experience building CI/CD pipelines and observability for production services (tracing, structured logging, metrics, alerting).
• Strong Python development skills with experience building production-grade services or tooling.
• Experience building LLM agents in production, including tool use, prompt and context design, and evaluation of agent behavior.
• Experience building Model Context Protocol (MCP) servers or comparable tool integrations that let LLM agents call internal APIs and data sources.
• Experience on a small team (roughly 3–10 engineers) where individuals own systems end to end, set their own priorities within a goal, and ship without a dedicated QA, SRE, or release function.
• Routine use of agentic coding harnesses (Claude Code, OpenCode, Codex, or equivalent) as part of day-to-day development, with a view on where they help and where they do not.
• Experience with Amazon Bedrock is a plus.
• Experience with Snowflake or SQL-based data access from services is a plus.
• Experience in a life sciences, chemistry, or drug discovery environment is a plus.
• Effective communicator who can work independently in a remote environment and collaborate with engineers, scientists, and IT security.

Primary Location

United States (Remote — working hours within one to two hours of US Eastern preferred; West Coast considered)

collaborate with both engineers and scientists.

Spectrix offers flexible schedules, competitive salaries, and a committed work environment to its employees; the company provides automatic benefits of life insurance and short and long term disability insurance. Optional benefits for full time employees include health, dental, and vision insurance, college loan assistance, and an employer matched retirement savings plan. Spectrix is not a placement agency.

Spectrix is participating in the E-Verify program of the U.S. Department of Homeland Security (phone number: 888-897-7781, and website: www.dhs.gov/E-Verify).

Pay: $48.00 - $58.00 per hour

Experience:

• Python: 7 years (Preferred)

Work Location: Remote