> Markdown version of [/jobs/ext/2834799-software-engineer-agentic-ai](https://www.wearedevelopers.com/jobs/ext/2834799-software-engineer-agentic-ai). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Software Engineer (Agentic AI

- **Company:** Innova Solutions
- **Location:** Columbus, OH, United States
- **Experience:** Expert
- **Salary:** $208,000.0 - $249,600.0
- **Contract:** Temporary to permanent
- **Skills:** Java (Programming Language), Application Programming Interfaces (APIs), Artificial Intelligence, Human-Computer Interaction, Python (Programming Language), Queueing Systems, RabbitMQ, Redis, Message Oriented Middleware, WebSocket, ReactJS, Large Language Models, Grafana, Multi-Agent Systems, Caching, Fastapi, Event Driven Architecture, Kubernetes, Low Latency, Performance Monitor, Apache Kafka, Virtual Agents, Microservices
- **Published:** September 10, 2026
- **Apply:** https://www.dice.com/job-detail/c629d4b5-6ae0-426c-9cd6-ff19be48e5f8

## About the Role

We are seeking a seasoned Senior Software Engineer (Agentic AI) to design, architect, and ship production-grade, multi-step autonomous AI agents and customer-facing agentic platforms. You will bridge the gap between complex LLM reasoning loops and high-scale backend engineering.
This is a hands-on builder role requiring demonstrated experience shipping real-world agents with real traffic metrics, robust context handoffs, and tool integration via MCP (Model Context Protocol) / custom agent harnesses.

 * Design, build, and deploy end-to-end multi-step autonomous AI agents and execution loops (ReAct, Plan-and-Execute) in production.
 * Architect resilient customer-facing agent platforms that handle unpredictable real-world inputs with low latency and tight guardrails.
 * Build Model Context Protocol (MCP) servers, custom agent harnesses, and tool interfaces for dynamic API execution.
 * Engineer state persistence, summarization layers, and memory engines for seamless context handoff between multi-agent networks and human workflows.
 * Develop active disambiguation loops, input schema validations (Pydantic), and fallback workflows to resolve vague or conflicting user prompts.
 * Scale agent microservices using asynchronous event-driven architectures, distributed message queues (Kafka/RabbitMQ), and Redis caching.
 * Implement real-time response streaming (SSE/WebSockets), prompt caching, and payload budgeting to minimize latency and token costs.
 * Integrate agent evaluation, step-by-step tracing, and performance monitoring using tools like LangSmith, Phoenix, or OpenTelemetry.
 * Proven track record of personally designing and shipping autonomous or multi-step AI agents into live production with real traffic metrics.
 * Demonstrated hands-on experience with customer-facing AI agent platforms, including real-time user interaction and edge-case handling.
 * Deep technical mastery of Model Context Protocol (MCP), agent harnesses, and orchestration frameworks (LangGraph, AutoGen, CrewAI, or custom engines).
 * 5+ years of core backend engineering expertise in Python (FastAPI, AsyncIO), Go, or Java within high-concurrency environments.
 * Practical knowledge of state management, context compression, short/long-term memory retrieval, and vector databases.
 * Expertise in microservices, asynchronous messaging platforms, distributed Redis caching, and event-driven architectures.
 * Hands-on familiarity with LLM observability tools, safety guardrails, and token cost/latency optimization strategies.