Embedding Retrieval and Agent Orchestration Engineer
Experience Level :
3–5 years About the Role :
We are seeking two experienced engineers to join our AI infrastructure team, focusing on embedding-based retrieval systems and agent orchestration frameworks. You will design, implement, and maintain components that enable intelligent agents to interact with APIs, external systems, and runtime environments using vector-based semantic search and secure orchestration logic. This role is ideal for engineers who thrive at the intersection of LLM infrastructure, search systems, and agentic automation. Key Responsibilities :
Vector Search & Retrieval :
Design and maintain vector databases (e.g., Pinecone, Weaviate, FAISS) for fast and scalable semantic search. Build and fine-tune embedding pipelines using models like OpenAI, Cohere, Hugging Face, etc. Implement and optimize retrieval-augmented generation (RAG) workflows for downstream LLM applications. Agent Frameworks & Orchestration :
Develop modular agent frameworks that support task delegation, contextual decision-making, and tool / API integrations. Orchestrate agent behaviors for multi-step tasks, leveraging state machines, workflows, or planning-based architectures. Integrate third-party and internal APIs with agents to enable dynamic task execution. Security and System Integration :
Ensure secure agent interactions with runtime environments, external APIs, and enterprise systems. Manage authentication, permissioning, and sandboxing where necessary to maintain data and system integrity. Required Qualifications :
3–5 years of experience in backend or ML engineering, with a focus on AI / ML infrastructure, LLMs, or intelligent agents. Strong knowledge of vector databases and embedding models for semantic search. Experience building or contributing to agentic systems (LangChain, LlamaIndex, AutoGPT, etc.) Solid programming skills in Python, with experience using frameworks like FastAPI, LangChain, or similar. Familiarity with API integration, secure data handling, and multi-system orchestration. Experience with cloud platforms (AWS, GCP, or Azure) and containerized environments (Docker, Kubernetes). Preferred Qualifications :
Hands-on experience with retrieval-augmented generation (RAG) systems. Exposure to knowledge graphs, task planning, or LLM tool use frameworks. Contributions to open-source agent frameworks or retrieval tools. Understanding of secure data access, API rate limiting, and runtime isolation.
#J-18808-Ljbffr
Engineer • Islamabad, Pakistan