AI Engineer
techcarrot FZ LLC
- Location
- Onsite (Dubai, Dubayy (Dubai))
- Employment
- Full-time
- Level
- Mid Level
Posted 5 days ago
About the role
techcarrot is a Dubai-based IT services provider specializing in digital transformation. This role involves designing, building, and deploying production-grade AI agents and LLM applications for enterprise clients.
Skills
Python
LLM
AI Agents
LangGraph
LangChain
RAG
Vector Databases
Pinecone
REST APIs
Docker
CI/CD
Orchestration
Conversational AI
Azure OpenAI
System Design
Observability
Full job details
We are seeking a hands-on AI Engineer to design, build, integrate and deploy intelligent AI agents for conversational and autonomous experiences. The role spans agent design, orchestration, tool integrations, evaluation, production deployment and continuous optimization. You will work closely with AI product managers, data scientists, backend engineers and business stakeholders to turn requirements into scalable AI solutions.
Key Responsibilities and Skills
AI Agent Development & Orchestration
- Design and deploy LLM-powered agents that reason, plan, retrieve information and execute actions.
- Build multi-agent systems with orchestrators, supervisors and specialized agents; implement tool calling, handoffs, delegation and multi-step execution.
- Develop context management, conversational memory, session management, guardrails, fallbacks and human-in-the-loop workflows.
- Develop both digital and voice enabled agents
- Integrate OpenAI, Azure OpenAI, Anthropic and other LLM providers.
- Deep understanding of speech models and their different architectures
- Support Arabic and English experiences.
- Optimize accuracy, latency, reliability and token costs.
RAG and Semantic Retrieval
- Develop system prompts, structured outputs and RAG solutions grounded in enterprise data.
- Build KB ingestion and retrieval pipelines
- Hands on experience with vector databases including pinecone and open search
Tool Development & Enterprise Integrations
- Build REST APIs, webhooks and Model Context Protocol (MCP) integrations for enterprise systems.
- Implement secure authentication, authorization and session handling.
- Enable business actions such as product discovery, recommendations, bookings and transactions.
Agent Evaluation, Testing & Optimization
- Build automated evaluations for task completion, tool selection, accuracy and response quality.
- Maintain test datasets for edge cases and multi-turn conversations; run regression tests.
- Debug execution traces, tool calls and context propagation; monitor production quality, latency and costs.
Production Deployment & Engineering Excellence
- Build maintainable production services with logging, monitoring, observability and tracing.
- Support CI/CD, code reviews, unit tests, documentation and secure development.
- Troubleshoot production incidents and integration or performance issues.
Technical Skills
- Strong hands-on Python, backend/API development and asynchronous programming.
- Proven delivery of LLM applications and AI agents, using at least one of LangGraph/LangChain, Kore.ai, OpenAI Agents SDK, Semantic Kernel or CrewAI.
- Strong understanding of orchestration, tool calling, context, memory and structured LLM workflows.
- Experience with REST APIs, microservices, RAG, embeddings and vector databases such as Pinecone or Azure AI Search.
- Working knowledge of Git, Docker, CI/CD, cloud deployment and LLM debugging.
- Voice and text conversational AI, STT/TTS and real-time architectures.
Requirements
Professional Experience
2+ years of software or AI engineering, including at least one production-grade agentic AI or LLM application.
Experience integrating AI solutions with enterprise systems and external APIs.
Strong independent problem solving and end-to-end delivery ownership.
Bachelor’s or Master’s degree in Computer Science, Software Engineering, AI or a related field.
Preferred Qualifications
Kore.ai Artemis or a similar enterprise agent platform.
MCP server/client development and agent-to-agent communication.
Azure cloud and enterprise AI deployment.
LLM evaluation and observability tools such as LangSmith or Langfuse.
Ideal Candidate
A builder, not just an AI enthusiast: someone who has personally written agent code, implemented tools, debugged LLM behavior and deployed AI applications. Comfortable owning orchestration, APIs, context management and production troubleshooting.