Job Description :
We are seeking for highly skilled Software Engineer with strong expertise in modern Python development and Large Language Model (LLM) ecosystems.
This role focuses on building scalable, production-grade AI systems, leveraging advanced API development, retrieval-augmented generation (RAG), and agentic frameworks.
You will work on cutting-edge AI solutions, contributing to the design, development, and deployment of intelligent systems within a distributed enterprise environment
Core Programming & Backend Development
-
Develop robust, scalable applications using Python (intermediate to advanced level)
-
Implement asynchronous programming patterns for high-performance systems
-
Design and build RESTful APIs using FastAPI
-
Write clean, maintainable, production-grade code
-
Develop and execute unit and integration tests
-
Debug and resolve issues in complex distributed systems
LLM Fundamentals & Prompt Engineering
-
Design and optimize prompts for various LLM use cases
-
Understand tokenization, context windows, and model limitations
-
Select appropriate models based on performance and cost trade-offs
-
Mitigate hallucinations and ensure grounded, reliable responses
Retrieval-Augmented Generation (RAG)
-
Build and maintain document ingestion and preprocessing pipelines
-
Implement chunking strategies (semantic, recursive, sliding window)
-
Generate and manage embeddings
-
Work with vector databases (e.g., pgvector)
-
Design hybrid search systems combining keyword (BM25) and semantic search
-
Optimize re-ranking and relevance tuning mechanisms
Agentic Frameworks & Orchestration
-
Design and implement multi-agent systems
-
Manage conversational and long-term memory
-
Build workflow orchestration pipelines for AI agents
Required Qualifications
-
Strong proficiency in Python with experience in asynchronous programming
-
Hands-on experience with FastAPI or similar frameworks
-
Experience building scalable backend systems
-
Solid understanding of LLM concepts and prompt engineering
-
Experience with RAG pipelines and vector databases (pgvector preferred)
-
Familiarity with distributed systems and debugging techniques
-
Experience with Docker and CI/CD pipelines