Generative AI Engineer

Company Description

Inlighten Technologies is a Silicon Valley company developing next-generation microLED technologies for augmented reality (AR) and AI-powered smart glasses. Our mission is to push the limits of microLED performance and manufacturing, enabling scalable, mass-market production. Join us to help create the displays that will power the future of immersive computing.

Role Description

As a Gen AI Engineer, you will design and deploy internal AI Agents supporting process engineering, manufacturing, quality, and operations teams. This role is focused on applied Generative AI engineering building high-quality RAG pipelines, agent reasoning workflows, evaluation frameworks, and domain-adapted generative AI solutions. Additionally, this role includes contributing to AI cybersecurity policy and governance for internal AI systems, ensuring responsible and secure deployment aligned with enterprise standards. This role offers growth opportunities into advanced agent architecture, multi-agent systems, and AI governance leadership.

Key Responsibilities

• Build production-grade RAG pipelines: ingestion, document parsing, chunking strategy, embeddings, retrieval tuning, and grounded answer generation (with iterative evaluation).
• Develop agent workflows for internal assistants (tool calling, task decomposition, multi-step workflows) using agent frameworks (e.g., LangChain/LangGraph-style orchestration).
• Integrate assistants with internal systems (examples: document management, engineering knowledge bases, ticketing, dashboards) via APIs and secure connectors; design for reliability and traceability.
• Improve response quality through structured evaluation methods: offline test sets, human review loops, hallucination analysis, and iterative prompt/system refinement.
• Own performance engineering: latency, cost/token controls, caching, batching, and retrieval optimization.
• Define and maintain AI usage policies for internal assistants: data handling rules, permitted sources, retention, and auditability (in partnership with IT/Security).
• Implement security controls and governance patterns: RBAC, encryption, audit logs, and privacy-by-design reviews for new data sources and tools.
• Support risk assessments and release checklists for agent tool integrations (secrets handling, logging hygiene, data leakage prevention).

Required qualifications

• Bachelor’s degree with 2+ years of industry experience, or Master’s degree with 1+ year of industry experience, in Computer Science, Engineering, or related field.
• Strong Python development skills and experience building backend services (e.g., FastAPI or similar)
• Demonstrated experience designing and deploying Retrieval-Augmented Generation (RAG) systems in production or production-like internal environments. (semantic chunking, retrieval tuning, evaluation strategies).
• Experience building tool-augmented LLM systems (e.g., function calling, standardized tool interfaces such as MCP, or similar architectures).
• Experience developing and implementing cybersecurity policies and governance standards for AI systems, including secure data management, access control frameworks, model risk mitigation, compliance alignment, and audit readiness

Preferred qualifications

• Master’s or PhD with 3+ years of experience working on applied AI, LLM systems, or advanced NLP solutions.
• Experience in designing and implementing MCP (Model Context Protocol) integrations, including structured context management, tool interface standardization, and extensible model interaction layers.
• Hands-on experience building Agent-to-Agent (A2A) or multi-agent systems, including orchestration patterns, inter-agent communication, tool-calling workflows, and multi-step reasoning pipelines.
• Experience fine-tuning or adapting LLMs for domain-specific applications.
• Experience with LLMOps/MLOps tooling (e.g., MLflow/CI-CD, automated tests, deploy/rollback) and evaluation harnesses.
• Familiarity deploying in restricted environments and/or higher-security settings for enterprise data.
• Semiconductor/manufacturing domain familiarity (MES, SPC, FDC systems).
• Mandarin language proficiency given collaboration with global engineering teams. (Nice-to-have; not mandatory.)