Job Title: Senior LLM Engineer

Location: On-site, Full-time
Employment: Full-time, Senior LLM / Applied NLP Engineer
Reports to: Technical Lead


Role summary

Design, build and maintain the LLM-driven assistants.
RAG pipelines, prompt engineering, retrieval stores (vector DB), safety/guardrails, and integrations with CRM to automate follow-ups and buyer interactions.

Core responsibilities
- Design RAG architecture: document ingestion, embeddings, vector DB selection and retrieval.
- Craft prompts, templates and system messages; implement adaptive prompting strategies and context windows.
- Integrate assistants with CRM endpoints and event streams to generate contextual drafts and buyer recommendations.
- Implement safety layers: hallucination detection, guardrails, confidence scoring and human-in-the-loop flows.
- Monitor assistant quality (response accuracy, latency, token usage), optimize for cost/performance.
- Build logs/audits and instrumentation for feedback loops (ratings, corrections) and iterative improvement.

Must-have (required)
- 3+ years applied NLP/LLM engineering or relevant ML engineering experience.
- Hands-on experience with RAG, vector DBs and at least one LLM provider (OpenAI/Bedrock/Anthropic).
- Strong Python skills and experience deploying model orchestration (SageMaker, LangChain, LlamaIndex or custom).
- Experience producing evaluation metrics for conversational agents and designing human-in-the-loop workflows.

Nice-to-have
- Prior experience in customer-facing assistants or CRM-integrated bots.
- Experience optimizing token usage & cost for high-volume production systems.
- Familiarity with on-prem or fine-tuning workflows for smaller LLMs.

If you are interested in working on cutting-edge projects in PropTech industry for our startup located on multiple countries inside GCC, feel free to send your CV to the following email: hr.dune.estates@gmail.com


➖➖➖➖➖


برای اطلاع از جديدترين فرصت های شغلی حوزه IT عضو کانال شوید
  👇
@workinja