Forward Deployed Engineer - GenAI

About the role

We are looking for a Forward Deployed Engineer specialized in Generative AI to design, build and ship production‑grade LLM‑powered applications for our clients in Saudi Arabia. You will turn business requirements into robust GenAI features such as copilot chat, enterprise search and Retrieval‑Augmented Generation pipelines.

Key responsibilities

• Develop GenAI applications including LLM‑driven features, chat‑copilot experiences and enterprise search.
• Design and implement RAG pipelines: chunking, embedding selection, hybrid retrieval, re‑ranking and GraphRAG.
• Fine‑tune models using LoRA/QLoRA and manage prompt versioning.
• Integrate LLM APIs (OpenAI, Anthropic, Azure OpenAI) and open‑source model endpoints with authentication, rate‑limiting and cost controls.
• Instrument applications for evaluation, logging, quality scoring and human‑feedback loops.
• Optimize latency and token cost through caching, batching and model routing.
• Translate client requirements into concrete feature specifications and communicate trade‑offs to non‑technical stakeholders.
• Collaborate with AI architects and data scientists, and document architecture and prompt design decisions.

Required profile

• 4–8 years of software engineering experience, including 1–3 years building GenAI/LLM applications.
• Strong Python development skills.
• Experience with LangChain, LlamaIndex or similar orchestration frameworks.
• Hands‑on knowledge of vector databases (Pinecone, Weaviate, pgvector) and graph databases (Neo4j).
• Understanding of LLM failure modes and mitigation strategies.
• Experience fine‑tuning models (LoRA/QLoRA) and using evaluation harnesses.
• Familiarity with enterprise GenAI platforms such as Azure AI, AWS Bedrock and Google Vertex AI.
• Ability to design APIs with authentication, rate‑limiting and streaming responses.
• Clear technical writing for architecture and prompt documentation.

Required skills

• Python
• LangChain
• LlamaIndex
• Pinecone
• Weaviate
• pgvector
• Neo4j
• LoRA / QLoRA
• Microsoft Azure AI
• AWS Bedrock
• Google Vertex AI
• OpenAI API
• Anthropic API
• Azure OpenAI
• API design
• Authentication & rate limiting
• Streaming responses
• Weights & Biases
• LangSmith