Forward Deployed Engineer - Agentforce Orchestration

Salesforce; Greater London; On-site; GBP 120,000 - 150,000; Full time

Job summary

Salesforce is seeking a Forward Deployed Engineer to design, build, and deploy autonomous AI solutions inside customer environments. You will work in a pod with a Deployment Strategist and Specialized AI Engineers to connect advanced agents to enterprise systems.

Key responsibilities include creating multi‑agent orchestration workflows, API-driven integrations, and real-time data grounding using RAG and vector stores. Travel 25–30% for on-site deployments is expected.

Qualifications

• 3+ years of software engineering or technical delivery experience, with at least one production AI, data engineering, or enterprise integration system deployed to production.
• Proficiency in Python, JavaScript/TypeScript, Java, or Apex.
• Hands-on AI grounding and LLM orchestration with RAG pipelines and vector indexes.

Responsibilities

• Design, construct, and deploy enterprise-grade AI solutions inside customer environments.
• Implement agent communication protocols and tool calling for multi-agent orchestration.
• Build real-time data grounding systems, knowledge bases, and ETL/ELT pipelines.
• Develop and maintain APIs (REST, GraphQL, webhooks) to connect AFO/Agentforce to enterprise systems.
• Write production code and deploy end-to-end workflows with humans in the loop.

Skills

Python

TypeScript

API design

LLM orchestration

MuleSoft

Data grounding

RAG pipelines

Agent protocols

Docker

Kubernetes

Salesforce ecosystem

Security governance

Tools

MuleSoft

Snowflake

Databricks

Terraform

Kafka

Job description

Salesforce is seeking a Forward Deployed Engineer to design, build, and deploy autonomous AI solutions inside customer environments. You will work in a pod with a Deployment Strategist and Specialized AI Engineers to connect advanced agents to enterprise systems.

Key responsibilities include creating multi‑agent orchestration workflows, API-driven integrations, and real-time data grounding using RAG and vector stores. Travel 25–30% for on-site deployments is expected.