Expert Agentic AI Engineer

Tkxel LLC; Lahore; Sur place; PKR 38 770 000 - 52 617 000; Plein temps

Résumé du poste

Tkxel is seeking an Expert AI Engineer to own the design, architecture, and deployment of agentic AI systems for a bilingual enterprise platform based on large language models. You will shape autonomous, multi-agent workflows and robust retrieval pipelines with production-grade RAG.

Responsibilities include architecting planning-driven systems, orchestration across agents and tools, and durable state management with retries.

Qualifications

• Deep, demonstrated expertise in agentic AI and LLM-based applications.
• Expert Python proficiency and strong software engineering foundation.
• Proven experience architecting agentic systems with planning and orchestration.
• Hands-on experience with LangGraph, LangChain, LlamaIndex, AutoGen, or CrewAI.
• Strong command of agent state management and tool calling standards (MCP).
• Experience designing DAG/state-machine workflows with durable state and retries.
• Experience building connectors to enterprise systems and OAuth-based auth.
• Production experience with background schedulers (Celery, Airflow, Temporal).
• Hands-on with vector databases and lexical search (BM25, Elasticsearch).
• Advanced prompt engineering, grounding, and evaluation strategies.

Responsabilités

• Architect agentic systems with planning, memory, and multi-step task execution.
• Define orchestration patterns across agents, tools, and LLM calls.
• Design robust single- and multi-agent architectures with state management.
• Develop the workflow layer with versioned, durable workflows and retries.
• Build connectors to enterprise content sources and APIs with secure auth.
• Own the background processing layer including ingestion and indexing.
• Oversee RAG layer including embedding, retrieval, and grounding generation.
• Implement and tune dense, sparse, hybrid, and metadata-based retrieval.
• Define prompts, grounding, and verification to ensure accurate responses.
• Set evaluation frameworks and quality gates to improve latency and reliability.

Connaissances

Agentic AI

LLM apps

Python

System design

Production deployment

LangGraph

LangChain

LlamaIndex

AutoGen

CrewAI

MCP

DAG/state machine

Celery

Airflow

Temporal

Milvus

FAISS

Qdrant

pgvector

BM25

Elasticsearch

RAG

Retrieval techniques

Prompt engineering

Grounding

Evaluation frameworks

Outils

Milvus

FAISS

Qdrant

pgvector

BM25

Elasticsearch

Description du poste

Tkxel is seeking an Expert AI Engineer to own the design, architecture, and deployment of agentic AI systems for a bilingual enterprise platform based on large language models. You will shape autonomous, multi-agent workflows and robust retrieval pipelines with production-grade RAG.

Responsibilities include architecting planning-driven systems, orchestration across agents and tools, and durable state management with retries.