AI/ML Architect

About the role

DevExplore is looking for an architect to build and deploy complex AI systems for our high-tech client base. You will bridge the gap between business needs and technical execution by designing end-to-end solutions centered on GCP and autonomous agent frameworks.

Key facts

What you'll do

• Build full-stack AI applications using Python-based backends and modern frontend frameworks like React, Angular, or Vue.
• Create system architectures that utilize GCP tools such as Vertex AI, BigQuery, Cloud Run, and Firestore.
• Integrate LLMs and agentic frameworks like LangChain, AutoGen, or CrewAI into production environments.
• Manage vector database setups and embedding workflows using tools like Pinecone, Weaviate, or FAISS.
• Lead technical workshops and architectural reviews to guide clients through cloud and AI transformations.
• Establish DevOps standards for microservices, CI/CD pipelines, and monitoring using GCP-native services.
• Translate client requirements into actionable technical roadmaps and mentor distributed engineering teams.

Requirements

• 10+ years of professional experience in software engineering and full-stack architecture.
• Proficiency in Python backend development using FastAPI, Flask, or Django.
• Hands-on history with agentic AI frameworks including LangChain, AutoGen, or CrewAI.
• Experience working with LLM APIs such as OpenAI, Claude, Gemini, or Mistral.
• Deep technical knowledge of the GCP ecosystem, specifically Vertex AI, Pub/Sub, Cloud Functions, and Cloud Storage.
• Practical experience with RAG pipelines and vector database management.
• Competence in containerization with Docker and Kubernetes, specifically GKE.
• Ability to work within Agile delivery frameworks and lead distributed teams.
• Understanding of REST, GraphQL, and API gateway design.

Skills & tools

• Python (FastAPI, Flask, Django)
• GCP (Vertex AI, BigQuery, Cloud Run, Cloud Build, Artifact Registry)
• AI Frameworks (LangChain, AutoGen, CrewAI)
• Frontend (React, Angular, Vue)
• Vector Databases (Pinecone, Weaviate, FAISS)
• Infrastructure (Docker, Kubernetes/GKE)