Forward Deployed Engineer

Meridian builds AI agents that complete real work for businesses, then stays to run them. Forward deployed engineers are the people who make that true on the ground.

You will join a client team, learn how a workflow actually runs, build and deploy agents for it on Prime, and keep those agents reliable once they are live. You will work directly with the people who do the job today, and you will bring what you learn back to the research team so the platform improves with every deployment.

What you will do

• Map how a client's work really gets done, including the exceptions that never reach a process document.
• Design, build and deploy agents for those workflows on Prime, from first version to live operation.
• Connect agents to the systems a client already uses: REST and GraphQL APIs, databases, queues, identity providers, ERPs and CRMs, and legacy tools that have no API.
• Build MCP servers and custom tools so agents can read from and act in client systems safely.
• Set up evaluations and review steps for every workflow, so quality is measured before and after go-live.
• Monitor agents in production, trace failures to their cause and fix them quickly.
• Handle client security needs: access scopes, secrets, audit trails and data residency.
• Work directly with client operators and leaders, and report progress in plain language.
• Bring field findings back to research and development, so recurring needs become platform capabilities.

What you bring

• Strong software engineering across backend services, integrations and data. You ship in Python or TypeScript and read other languages with ease.
• Experience building agents with current tools: the Claude Agent SDK, the OpenAI Agents SDK, LangGraph or similar, tool calling, structured outputs and MCP.
• Experience with retrieval over messy business data: documents, spreadsheets, email and databases, with embeddings, chunking and reranking done well.
• Sound judgment about where language models fail, and the habit of testing for it with evaluation sets and tracing tools.
• Production fundamentals: Postgres, REST design, webhooks, queues, Docker, CI/CD and a major cloud such as AWS, Azure or GCP.
• Experience with authentication and integration standards such as OAuth, SSO and SAML.
• Comfort working with clients: you listen first, ask good questions and own the outcome.
• The ability to move from an unclear problem to a working system without waiting for a spec.
• Clear written and spoken communication.

Strong advantages

• Experience in consulting, solutions engineering or a similar embedded role.
• Experience with voice agents and real-time pipelines, using tools such as LiveKit, Twilio or the OpenAI Realtime API.
• Experience with browser and computer-use agents, and with automation tools such as Playwright for systems that have no API.
• Experience with enterprise platforms such as Salesforce, HubSpot, SAP, Oracle NetSuite or Microsoft Dynamics.
• Experience with workflow engines such as Temporal or n8n.
• Experience with OpenTelemetry and AI observability tools.
• Arabic, or experience delivering projects across the Middle East.

How we hire

We keep the process short: an introductory conversation, a practical exercise based on real work, and time with the people you would work alongside.

About Meridian

Meridian builds and operates AI agents for businesses that run on people and process. We map how the work gets done, deploy agents that complete it end to end under human supervision, and stay to run them in production.