Job Title: AI Engineering & Agentic Systems
Work Location: Eden PrairieMN55344
Vendor Rate: 71.82 - 108.00
Contract duration: 6
Target Start Date: 29 Jun 2026
Does this position require Visa independent candidates only? YES
**hybrid work set-up**
Job Details:
Must Have Skills
Experience with agent and LLM ecosystem tools Google Agent Development Kit (ADK), LangChain & LangGraph (agent orchestration), Model Context Protocol (MCP) FastMCP or similar connector development, A2A ACP interagent communication protocols
Proficiency with LLM streaming APIs Vertex AI Gemini, AWS Bedrock, OpenAI
Familiarity with OASF (Open Agentic Schema Framework) agent schema and registry patterns
Nice to have skills
Detailed Job Description
Hands-on production experience building LLM-powered applications or agentic systems this is not a traditional ML/data science role (no model training, no heavy ML pipelines)
Strong understanding of multi-agent orchestration, tool-using agents, Retrieval-Augmented Generation (RAG), structured outputs, function calling, and Human-ON-the-Loop (HOTL) workflows
Experience with agent and LLM ecosystem tools: Google Agent Development Kit (ADK), LangChain & LangGraph (agent orchestration), Model Context Protocol (MCP) FastMCP or similar connector development, A2A / ACP inter-agent communication protocols
Proficiency with LLM streaming APIs: Vertex AI / Gemini, AWS Bedrock, OpenAI
Familiarity with OASF (Open Agentic Schema Framework) agent schema and registry patterns
Core Languages & Frameworks
Python strong production experience (primary language required)
TypeScript / JavaScript good to have
APIs, Services & Integration
FastAPI / AsyncIO REST API design, webhooks, event-driven services
OpenAPI / AsyncAPI / Protobuf API contract design
Apache Kafka, GCP Pub/Sub event streaming and async agent communication
Testing & Quality Engineering
Automated Test-Driven Development (TDD) designing systems with test-first discipline
Regression testing ensuring behavioral stability across rapid iterations
End-to-End (E2E) testing validating agent workflows across services and integrations
Test automation for APIs, agents, and event-driven systems
Platform, Infrastructure & Cloud
Experience working in cloud environments (GCP preferred, AWS)
Kubernetes; Google Cloud Run / Cloud Run Jobs hands-on operational depth
Docker containerization
GitHub Actions, Cloud Build CI/CD pipelines
Familiarity with microservices, distributed systems, and Infrastructure-as-Code (Terraform, etc.)
Data & Storage
VectorDB retrieval systems for RAG and knowledge grounding
Firestore, MongoDB, or equivalent NoSQL
PostgreSQL / SQL relational databases
Google Cloud Storage (GCS) artifact and deployment package management
Redis caching
Observability & Reliability
OpenTelemetry tracing, spans, structured observability
Grafana dashboards and SLO visualization
DORA metrics & SLO engineering
Security, Identity & Governance
Open Policy Agent (OPA) policy enforcement in agent workflows
SPIFFE / Workload Identity non-human identity and mTLS
Mindset & Work Style
Genuinely hands-on, strategic AI-first mindset engineer who takes full ownership of work
Thrives in a fast-paced environment with continuous experimentation
Actively leverages modern AI-assisted development tools GitHub Copilot, Codex, and Claude
Track record of shipping production-grade systems, not prototypes
Comfortable with ambiguity and rapid evolution of AI tooling
Preferred Skills and Attributes
Experience with prompt/version management and evaluation tooling (2+ years)
Skills generation and agent builder experience
Familiarity with emerging agent frameworks and orchestration patterns
Understanding of AI observability and evaluation frameworks (quality, latency, cost, safety)
Experience with Responsible AI practices (guardrails, safety, auditability)
Knowledge of cost/performance tradeoffs in LLM systems
Experience building monitoring, logging, and feedback loops for AI systems
Mentoring experience ability to guide engineers on AI-first development approaches
Experience contributing to platform-first abstractions that enable other engineers to build AI features
Familiarity with closed-loop workflows (detect reason act validate)
Primary Responsibilities
1. Build AI-Native Capabilities Design and implement agentic workflows and multi-agent systems that solve real business problems across operations, service health, and enterprise workflows. Develop LLM-powered features using APIs (OpenAI, Google, AWS, Anthropic, etc.) with patterns such as RAG, tool use, planning, and memory. Translate business problems into composable AI capabilities, not one-off solutions.
2. Contribute to the AI Delivery Platform (AIDLC) Build reusable components across platform layers including prompt orchestration, agent frameworks, tooling/API integration layers, evaluation, guardrails, and observability. Help define and standardize AI development patterns, templates, and accelerators. Enable other engineers to build AI features through platform-first abstractions.
3. Deliver End-to-End AI Features Own delivery from concept prototype production. Implement closed-loop workflows (detect reason act validate). Integrate with enterprise systems via APIs, event streams, and observability platforms.
4. Operationalize AI at Scale Implement evaluation frameworks (quality, latency, cost, safety). Build monitoring, logging, and feedback loops for AI systems. Ensure solutions meet enterprise standards for governance, auditability, and reliability.
5. Drive AI Engineering Excellence Apply modern best practices in prompt engineering and versioning, agent orchestration and tool use, retrieval strategies and knowledge grounding. Mentor engineers on AI-first development approaches. Contribute to a culture of rapid experimentation and measurable delivery.
Prior Experience, Industry Background, or Domain Expertise
5+ years in software/platform engineering with a strong delivery focus
Prior experience building production-grade LLM-powered applications or agentic systems (not experimental/prototype-only)
Background in enterprise platform engineering, cloud-native development, or distributed systems
Experience with healthcare, insurance, or regulated industry environments is a plus
Familiarity with enterprise AI delivery lifecycle concepts governed, scalable, auditable AI systems
Understanding that this role is fundamentally different from traditional roles:
o Not a data scientist no model training, no heavy ML pipelines
o Not a one-off builder contributing to a shared platform
o Not experimental-only production delivery at scale
o Not automation-only intelligent, reasoning systems, not scripts
AI Skills
All contractor resources are expected to demonstrate baseline proficiency in enterprise-approved AI tools as part of their day-to-day responsibilities. This includes, but is not limited to:
Consistent Use: Maintain a minimum of 90% weekly usage of AI tools such as GitHub Copilot, Microsoft 365 Copilot, and other GenAI platforms approved by the enterprise.
Applied Productivity: Leverage AI tools to enhance coding, documentation, data analysis, and decision-making workflows.
Continuous Learning: Stay current with evolving AI capabilities and features, and apply them to improve delivery quality and velocity.
We are seeking candidates who actively leverage GitHub Copilot, Codex, and Claude as core development tools not as optional add-ons. AI-assisted development is a baseline expectation, not a differentiator.
Optional Fields:
AI Delivery Lifecycle Platform (AIDLC) An enterprise platform enabling rapid, scalable, and governed delivery of AI-powered capabilities across the organization. Provides standardized patterns for agent orchestration, LLM integration, evaluation, observability, and lifecycle management.
Agentic Workflow Development Multi-agent systems and intelligent workflows for operations, service health, and enterprise use cases
Platform Reusable Component Library Prompt orchestration frameworks, tooling/API integration layers, evaluation and guardrails, observability pipelines
Please share any additional information that would help vendors better understand the role, candidate profile, or hiring priorities.
We are seeking a genuinely hands-on, strategic AI-First Mindset Engineer who takes full ownership of work, thrives in a fast-paced environment, and continuously experiments. Please ensure candidates are thoroughly vetted against these expectations:
Candidates must demonstrate real production experience with agentic AI systems not just academic or prototype exposure
Look for evidence of shipping production-grade systems end-to-end (concept deploy operate)
Verify active, daily use of AI-assisted development tools (GitHub Copilot, Codex, Claude)
This is a platform engineering role candidates should show experience building for reuse and scale, not one-off solutions
Strong preference for candidates who think in AI capabilities and workflows, not just code
Candidates should balance experimentation with production discipline and be focused on business impact, not just technical novelty
Minimum years of experience
>10 years
Certifications Needed :No
Top 3 responsibilities you would expect the Subcon to shoulder and execute
Build AINative Capabilities Design and implement agentic workflows and multiagent systems that solve real business problems across operations, service health, and enterprise workflows. Develop LLMpowered features using APIs OpenAI, Google, AWS
Contribute to the AI Delivery Platform AIDLC Build reusable components across platform layers including prompt orchestration, agent frameworks, toolingAPI integration layers, evaluation, guardrails, and observability
. Deliver EndtoEnd AI Features Own d
Interview Process (Is face to face required?)
No
Any additional information you would like to share about the project specs/ nature of work
Project Code: Child code for DNA