Responsibilitie

• Deliver high quality Python services and AI components. 
• Take part in team ceremonies and collaborate closely with engineers and product. 
• Understand product requirements and propose simple, practical designs. 
• Implement clean, testable, maintainable Python code. 
• Build retrieval pipelines for embeddings, chunking, indexing, and vector search. 
• Integrate LLMs using APIs or local models served via Ollama or similar. 
• Add automated tests, from unit to integration. 
• Diagnose and fix production issues. 
• Improve system performance, reliability, and cost efficiency. 
• Contribute to code reviews and maintain coding standards.
• Work with DevOps, product, QA, and design to deliver reliable AI features.

Knowledge, Skills and Experience Required

Proven experience required: 

• Clear and accurate technical communication in English. 
• Can write performant, testable and maintainable Python code  
• 6+ years of proven commercial Python experience 
• 2+ years working directly with AI systems, LLMs, retrieval or model integration. 
• SQLite and JSON storage and Retrieval 
• Code UI in Streamlit or similar frameworks 
• Use LangChain pragmatically for orchestration, tools, retrievers, and evals. 
• Experience serving quantised models using Ollama, vLLM, or similar. 
• Experience building backend services and APIs. 
• Familiarity with LLMs, tokenisation, prompting, and model serving. 
• Experience with retrieval systems, embeddings, and vector databases. 
• Build AI data ingestion and indexing pipelines. 
• Build custom Docker containers for various combinations of code/tool of Python/SQLite/Ollama/FastAPI or other similar  tools 
• Containers (Docker) and Kubernetes, one major cloud, AWS or Azure. 
• Model relationships with graph databases, Memgraph, Neo4j or Neptune, and blend graph plus vector search.
•  Understanding of Event-based framework (Pub/Sub, Queues) 

General Experience Required:  

• Knowledge of Pyventus or similar python based events framework 
• Knowledge of RAG evaluation frameworks. 
• ML in Python (will consider other ML experience as well) 
• Experience with monitoring and tracing, for example OpenTelemetry, Prometheus, or Grafana.

Nice to have

• Agentic framework (Claude skills, MCP-server/client, other similar frameworks) 
• PortKey or similar AI Gateways
• Langfuse or similar LLM monitoring tools

Working Conditions

Hybrid work model with four days working from the Kochi office and Fridays as work-from-home.
Employees can choose from multiple shift options : 9:00 am to 6:00 pm, 10:00 am to 7:00 pm and 11:00 am to 8:00 pm