At Saman Electronic Payment (SEP), we are building the next generation of internal chatbots powered by Large Language Models (LLMs) — systems that not only answer general questions but also securely retrieve and process organizational data.

Role Overview:
We are seeking an NLP/LLM specialist to play a key role in designing and developing intelligent chatbots with Retrieval-Augmented Generation (RAG) and secure integration into internal systems.

Responsibilities:

• Design and develop intelligent chatbots using LLMs, RAG, and tool/function calling
• Manage and optimize data pipelines: ETL, chunking, embedding, hybrid search, and reranking
• Optimize models for accuracy, cost, and latency (LoRA/QLoRA, quantization, speculative decoding)
• Implement traceability and debugging mechanisms (latency monitoring, structured logging, version control for prompts and models)
• Ensure system safety
• Collaborate with DevOps/MLOps teams for deployment and observability

Requirements:

• Proficiency in Python and Git; solid engineering habits (testing, packaging, code review)
• Practical experience with RAG (chunking, embeddings, hybrid search, rerankers)
• Experience with LLM applications (prompt design, agents, state machines)
• Familiarity with vector databases (Weaviate, Milvus, pgvector)
• Experience with Docker and Kubernetes

Nice to Have:

• Experience with ASR/TTS or OCR pipelines
• Knowledge of knowledge graphs or learning-to-rank approaches
• Proven experience in deploying LLM or RAG systems in production
• Advanced degree (MSc/PhD) in Computer Science or related fields