About the role
Leads the development of complex AI-powered software features and contributes to architectural decisions across Tekion's T1 and ORBIT platforms. Brings deep Java engineering expertise combined with hands-on experience in agentic orchestration, LLM pipelines, and graph/vector data systems. Mentors junior engineers, promotes engineering best practices, and drives technical excellence across design, implementation, and production operations.
Responsibilities:
- Design and build scalable, high-performance Java services and agentic microservices powering Tekion's AI platform.
- Develop and integrate LLM-based agent pipelines using LangChain, LangGraph, and LangFlow from prompt design through production deployment.
- Implement vector-based semantic search and knowledge retrieval using Qdrant, and model complex domain relationships using Neo4j graph databases.
- Build and maintain durable, fault-tolerant async workflows using Temporal, ensuring reliable execution of long-running agent orchestration tasks.
- Architect and implement multi-agent orchestration patterns, routing, tool use, memory management, and agent chaining across conversational and batch workloads.
- Establish and maintain observability standards: distributed tracing, structured logging, LLM call telemetry, latency dashboards, and cost-per-call monitoring.
- Lead feature development end-to-end from technical design through code review, QA sign-off, and production rollout.
- Collaborate with architects and senior engineers to influence system design and platform strategy.
- Mentor junior and mid-level engineers; conduct rigorous code reviews; drive adoption of engineering best practices across the team.
- Proactively identify performance bottlenecks, security gaps, and reliability risks in development and production environments.
- Champion quality, contribute to test strategy, promote TDD/BDD practices, and ensure adequate coverage across unit, integration, and contract tests.
Requirements:
- 5-7 years of professional software engineering experience with Java as the primary language.
- Hands-on experience building or integrating LLM-based pipelines using LangChain, LangGraph, or LangFlow in production or near-production environments.
- Working knowledge of Qdrant or a comparable vector database for embedding-based semantic search and retrieval.
- Experience with Neo4j or a graph database for modelling entity relationships and knowledge graphs.
- Practical experience with Temporal (or a comparable durable execution engine such as Conductor or Cadence) for workflow orchestration.
- Strong understanding of agentic orchestration patterns: tool use, ReAct loops, multi-agent routing, memory, and context management.
- Solid grasp of observability fundamentals: distributed tracing (OpenTelemetry), structured logging, metrics, and alerting in production systems.
- Quality-first mindset: experience defining test strategies, writing testable code, and driving coverage across unit, integration, and e2e layers.
- Bachelor's or Master's degree in Computer Science, Engineering, or a related technical field.
- Strong communication skills; able to articulate technical trade-offs to both engineering and non-engineering stakeholders.
Preferred:
- Experience deploying AI-native services on cloud platforms (AWS, GCP, or Azure) with container orchestration (Docker + Kubernetes).
- Familiarity with LLM token cost management, context window optimisation, and prompt engineering at scale.
- Exposure to enterprise SaaS platforms or automotive retail technology.
- Knowledge of security best practices for LLM-powered systems: prompt injection, output validation, and PII handling.