Description

Key Responsibilities

• Design and build AI-powered features using large language models and related tooling
• Develop and maintain prompt architectures that drive consistent, high-quality outputs
• Implement retrieval-augmented generation pipelines using enterprise data sources
• Build and orchestrate agent-based workflows to automate targeted tasks

• Integrate LLM APIs such as Anthropic Claude and OpenAI into production systems
• Design context management strategies to ensure outputs are grounded, relevant, and accurate
• Manage tradeoffs across latency, cost, and performance in AI workflows
• Continuously improve system behavior through prompt iteration and architecture refinement

• Partner with Software Engineers to integrate AI capabilities into applications, APIs, and user interfaces
• Align with the Lead Engineer on technical direction, architecture, and implementation decisions
• Work with QA Engineers to define evaluation criteria, testing strategies, and quality thresholds for AI outputs
• Translate product requirements into scalable, production-ready AI solutions

• Define and implement approaches for evaluating non-deterministic AI outputs
• Build test cases, benchmarks, and evaluation pipelines to track output quality over time
• Identify failure modes and iterate on prompts, pipelines, and orchestration logic
• Ensure consistency and reliability as models, prompts, and data sources evolve

• Stay current with advancements in LLMs, vector databases, and agent frameworks
• Experiment with new tools and techniques to improve speed, quality, and capability
• Contribute reusable patterns, components, and best practices across pods

Skills, Knowledge and Expertise

• 4+ years of experience in software engineering, applied AI, or machine learning development
• Strong programming skills in Python and/or JavaScript
• Hands-on experience working with LLM APIs such as Anthropic Claude, OpenAI, or similar
• Experience designing and implementing prompt architectures and prompt engineering techniques
• Experience building retrieval-augmented generation pipelines and working with vector databases
• Familiarity with agent orchestration frameworks and multi-step AI workflows
• Experience integrating AI capabilities into applications via APIs and backend systems
• Strong understanding of handling structured and unstructured data in AI systems
• Ability to evaluate, debug, and improve non-deterministic AI outputs
• Experience working in a fast-paced, product-oriented development environment
• Strong problem-solving skills and ability to operate in ambiguous, evolving contexts
• Ability to collaborate closely with engineers, product managers, and QA within a pod structure
• Excellent communication skills and ability to explain technical concepts clearly
• Curiosity and willingness to learn domain-specific workflows, particularly within engineering and AEC contexts

About Trilon Group

Our hiring process