SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Replicant builds an AI platform for contact centers that uses large language models and conversation intelligence to resolve customer requests, improve agent performance, and handle millions of calls monthly for Fortune 500 companies and high-growth innovators.
As a Support Software Engineer on the Technical Support team, you own the post-launch reliability and enhancement of AI agents and products once they're live in production. This is an investigative and fixing role where you serve as an escalation point for frontline support engineers, working directly with customers on live issues.
Key responsibilities:
- Own bugs, issues, and enhancement requests for bots and products in production, working escalations that frontline support cannot resolve
- Investigate root cause across logs, code, and internal tooling, then design, implement, test, and ship fixes using agentic coding tools like Claude Code and Cursor
- Take a holistic view of platform health spanning AI Agent behavior, Conversation Intelligence, telephony, SSO/authentication, integrations, API issues, LLM prompting, and legacy TypeScript SDK bots
- Serve as primary point of contact for customers on live issues: gather requirements, confirm reproduction steps, and coordinate testing
- Act as escalation point for frontline Support Engineers, providing technical context and guidance
- Build and maintain internal tools that accelerate future investigations and fixes
- Gather customer feedback for Product team and adhere to team SLAs
You'll work on real production traffic with genuine autonomy on a small team where your work is visible to customers and leadership. You are measured on reliability and resolution, not features shipped.
Requirements:
- Debugging depth: comfortable reading unfamiliar code, tracing logs, and reasoning across a broad surface area including production debugging, distributed systems, REST APIs, and authentication (OAuth 2.0)
- Agentic fluency: use agentic coding tools (Claude Code, Cursor, or equivalent) as your default investigation and fixing method; comfortable maintaining legacy TypeScript code
- Customer-facing strength: gather requirements directly from customers, confirm reproduction steps, coordinate tests, and explain fixes in plain language, including with frustrated customers
- Nice to have: experience with LLM prompting or debugging conversational AI/voice systems; experience with enterprise tools like Salesforce, Zendesk, or Jira; experience building internal tooling or automations that reduced manual engineering or support load
