Agentic AI Architect

We are looking for a research-oriented Agentic AI Engineer / AI Solutions Architect who actively explores emerging AI technologies and converts promising ideas into scalable, production-ready Agentic AI solutions.
The ideal candidate should be comfortable working across:
Research → Experimentation → Evaluation → Production
Key Responsibilities
Research and experiment with new LLM, GenAI and Agentic AI models, tools, frameworks and architectures.
Explore and validate new AI approaches to identify practical enterprise use cases.
Design Agentic AI solutions using patterns such as ReAct, Planner-Executor, Supervisor/Worker and Multi-Agent systems.
Build solutions involving RAG, tool/function calling, agent memory, state management and workflow orchestration.
Evaluate different models and approaches based on accuracy, performance, cost, scalability and production feasibility.
Take validated AI approaches into production, including architecture, deployment, monitoring and optimization.
Stay current with AI research, open-source developments and emerging Agentic AI technologies and identify their potential enterprise applications.
Required Skills
5+ years of software/technology experience with 3+ years in AI/ML, GenAI or LLM-based solutions.
Strong research and experimentation mindset with hands-on AI/Agentic AI projects.
Strong Python skills.
Experience with LangGraph, LangChain, Semantic Kernel, AutoGen, CrewAI or similar frameworks.
Strong understanding of LLMs, RAG, embeddings, vector databases, prompt engineering and tool calling.
Experience deploying AI solutions on AWS, Azure or GCP.
Ability to move from new AI ideas and experimentation to production implementation.
Preferred
Experience in AI R&D / Applied AI / Research Engineering.
Experience with AI evaluation and observability tools such as Ragas, DeepEval or Promptfoo.
Research papers, open-source contributions, technical projects or strong GitHub work are a plus.
Knowledge of AI security, MCP, multimodal AI, Docker/Kubernetes is a plus.
What We Are Looking For
Someone who doesn't just implement existing AI solutions, but continuously asks:
“What is new, how can we experiment with it, does it solve the problem better, and can we take it to production?”

Work arrangement
Hybrid