Artificial Intelligence Engineering Lead

Provoke Solutions; United States; Remote; USD 83,000 - 138,000; Full time

Benefits offered by this job

Remote work (US)

Competitive compensation

Job summary

Provoke Solutions is seeking a Software Engineer to join the AI Infrastructure / Build Team. You will help build the platform that enables engineering teams to integrate AI agents into GitLab workflows, focusing on event-driven services and Bedrock integration within CI/CD pipelines.

This hands-on role requires strong Python and AWS development experience, plus practical work integrating large language models into production applications.

Qualifications

• 3–5 years of professional software engineering experience.
• Strong Python development skills with AWS SDK (Boto3).
• Hands-on experience with AWS services and production-grade Lambda functions.
• Experience integrating LLM APIs into applications.

Responsibilities

• Build and maintain AWS Lambda functions that orchestrate AI agents and Bedrock invocation.
• Develop event-driven architectures using SQS, EventBridge, and other AWS services.
• Create modular platform components to support new agents and tools.
• Integrate AI agents into GitLab CI/CD pipelines.
• Build pipeline stages with context for branches, diffs, and tests.

Skills

Python

AWS

Boto3

CI/CD

GitLab CI/CD

LLM integration

Security best practices

Education

Bachelor's degree in CS or related field

Tools

Lambda

API Gateway

SQS

EventBridge

IAM

Secrets Manager

CloudWatch

Job description

Provoke is a global consulting firm building AI-native solutions that transform how work gets done. Founded on a culture of innovation, growth, and curiosity, we partner with global clients to design and deploy agentic AI embedded directly into workflows so teams move faster, think smarter, and scale with purpose.

Overview

As a Software Engineer on the AI Infrastructure / Build Team, you will help build and extend the platform that enables engineering teams to integrate AI agents into their GitLab workflows. You will develop event-driven services, integrate Amazon Bedrock with CI/CD pipelines, implement security and observability controls, and build reusable tools and capabilities that teams can adopt without redesigning their existing systems.

This is a hands-on engineering role for someone who combines strong Python and AWS development skills with practical experience integrating large language models (LLMs) into production applications.

The ideal candidate brings a blend of technical AI expertise, business acumen, and cross-functional collaboration skills, along with the ability to independently evaluate complex problems and drive transformation initiatives from opportunity identification through solution design.

Key Responsibilities

Agent Orchestration and Platform Development

• Build and maintain AWS Lambda functions that orchestrate AI agents, including routing, queue management, and Amazon Bedrock Agent invocation.
• Develop event-driven architectures using Amazon SQS, EventBridge, and other AWS services.
• Create modular, reusable platform components that support new agents, tools, and triggers without requiring changes to unrelated services.
• Integrate AI agent capabilities into GitLab CI/CD pipelines using .gitlab-ci.yml configurations.
• Build pipeline stages that provide agents with branch, code-diff, and test context.
• Enable agents to generate artifacts, return execution results, and determine whether downstream pipeline stages proceed based on defined success or failure criteria.

Agent Tools and Knowledge Integration

• Develop reusable, AWS Lambda-backed Action Groups that allow agents to interact with approved tools and services.
• Configure Amazon Bedrock Knowledge Bases using Amazon S3 and Amazon OpenSearch Service to provide agents with relevant organizational context.
• Design extensible integrations that support shared knowledge, retrieval, and agent capabilities across engineering teams.

Security and Governance

• Implement Amazon Bedrock Guardrails, including input and output filtering, sensitive-data protection, and content restrictions.
• Apply least-privilege IAM policies, secure secrets management, input validation, and agent-specific permission boundaries.
• Identify and mitigate risks associated with prompt injection, unauthorized tool access, and sensitive-data exposure in LLM-powered workflows.

Cost Optimization and Performance

• Implement token and cost management controls, including model selection, intelligent routing, semantic caching, and context-window management.
• Contribute to an LLM gateway that supports efficient model usage and cost-aware execution.
• Monitor usage patterns and identify opportunities to improve platform performance and resource efficiency.

Observability and Reliability

• Build CloudWatch dashboards, structured logging, metrics, and tracing capabilities.
• Maintain audit trails for agent executions, including inputs, outputs, decisions, token consumption, and associated costs, with appropriate security and data-retention controls.
• Troubleshoot production issues and improve the reliability, debuggability, and transparency of agent workflows.

Required Skills & Experience

• 3–5 years of professional software engineering experience.
• Strong Python development skills, including experience using the AWS SDK for Python (Boto3) to build production-grade Lambda functions and event-driven services.
• Hands-on experience with AWS services, including Lambda, API Gateway, SQS, EventBridge, IAM, Secrets Manager, and CloudWatch.
• Practical experience with Amazon Bedrock, specifically Agents, Action Groups, Knowledge Bases, and Guardrails.
• Experience integrating LLM APIs into applications, such as Amazon Bedrock, OpenAI, Anthropic, or comparable platforms.
• Familiarity with CI/CD platforms, preferably GitLab CI/CD, including pipeline configuration and automation.
• Understanding of cloud security principles, including least-privilege access, secure credential handling, input validation, and data protection.
• Awareness of LLM-specific security risks, including prompt injection and potential data leakage.
• Experience implementing observability practices such as structured logging, metrics, tracing, and production troubleshooting.
• Ability to develop modular, maintainable services with clear interfaces and well-defined responsibilities.

Desirable Qualifications

• Experience with application security and code analysis tools, including SAST, DAST, or SCA.
• Experience building internal developer platforms, engineering tools, or shared services used by multiple development teams.
• Familiarity with Amazon Bedrock AgentCore or comparable agent development and runtime frameworks.
• Experience implementing LLM gateways, model routing, semantic caching, or token-usage optimization.
• Understanding of retrieval-augmented generation (RAG), knowledge retrieval, and enterprise AI integrations.
• Experience designing secure, scalable, event-driven architectures in AWS.

Working Hours

US time zone alignment. Flexibility to accommodate client schedules is part of the role.

Location

Remote (United States)

Role Type & Work Authorization

This role is offered as W2 hourly employment. Candidates must be authorized to work in the U.S. without VISA sponsorship.