Role Description As a Backend Engineer at Freehand, you will build the distributed systems, APIs, and AI agent infrastructure that power AI-driven enterprise execution. You will design and scale backend services that orchestrate agents, enforce business rules, integrate deeply with ERPs and financial systems, and ensure that every action taken by AI is secure, traceable, and reliable. This is not CRUD SaaS work. You will be building the backend for autonomous systems handling real money and real risk—and the observability and testing infrastructure to keep those systems trustworthy at scale. What You Will Build - High-throughput, low-latency backend services that power AI agent execution - Workflow orchestration systems for complex, multi-step enterprise processes - Rule engines and policy enforcement layers for enterprise compliance - APIs and integrations with ERPs, procurement tools, payment rails, and data providers - Observability, audit, and traceability systems for AI-driven decisions - Secure data pipelines handling sensitive financial and operational data - Agent execution runtimes, multi-agent coordination, and observability pipelines for AI-driven workflows - Evaluation and testing harnesses to validate agent behaviour, reliability, and output quality Key Responsibilities - Backend & Systems Engineering - Design and build scalable backend services using modern system design principles - Own critical services that manage procurement, invoice, contract, and payment workflows - Implement orchestration logic that coordinates AI agents, humans, and enterprise systems - Build resilient systems with strong guarantees around idempotency, retries, and failure handling - API & Integration Development - Build and maintain robust APIs consumed by AI services, frontend apps, and external systems - Integrate deeply with ERP systems, procurement platforms, and payment gateways and financial rails - Handle complex data normalisation and transformation across enterprise systems - AI Agent Building, Observability & Testing - Build agent execution runtimes: tool calling, context management, memory, and multi-step reasoning loops - Implement multi-agent coordination—parallel and sequential workflows with human-in-the-loop escalation - Instrument agent execution with full trace capture: tool calls, LLM I/O, latency, token usage, and cost - Design evaluation frameworks for agent output quality, task success rate, and regression detection - Build automated test harnesses for agent pipelines—unit, integration, and replay-based regression tests - Define and track agent reliability metrics: task completion rate, escalation rate, cost per workflow, and SLA adherence - Reliability, Security & Compliance - Design systems with enterprise-grade reliability, monitoring, and alerting - Implement fine-grained access controls, audit logs, and data security best practices - Ensure backend systems meet compliance requirements for financial and regulated data - Performance & Scale - Optimize backend systems for latency, throughput, and cost efficiency - Design services that scale across global customers and billions of transactions - Proactively identify and fix bottlenecks in distributed systems Qualifications - Strong experience building production backend systems - Strong proficiency in Python, NodeJS, Go, or similar backend languages - Deep understanding of distributed systems, APIs and microservices, databases (SQL and NoSQL), and message queues / event-driven architectures - Proven experience designing scalable, fault-tolerant systems - Strong grasp of consistency, concurrency, and data integrity trade-offs - Experience working with financial systems, enterprise SaaS platforms, or sensitive/regulated data Bonus (Nice to Have) - Experience with multi-agent orchestration and agent-to-agent communication protocols - Familiarity with event sourcing, rule engines, or policy systems - Knowledge of production debugging at scale and observability tooling - Contributions to or experience with open-source LLM tooling ecosystems - Experience with agent observability tooling (LangSmith, Arize, Helicone, or custom trace pipelines) and eval design - Experience building LLM-powered systems in production; familiarity with agent frameworks (LangChain, CrewAI, or custom) Benefits - Be at the leading edge of the gen AI-native wave—replacing traditional software with intelligent agents. - Work with a team that's shipping AI into production at some of the world's most demanding enterprises. This is not a small pilot—it's global scale in production. - Take ownership of a vertical that will redefine how enterprises operate—replacing entire human teams and displacing entire categories of SaaS. - Join a culture that prizes clarity, velocity, and ambition.