Senior AI/ML Engineer

Milestone-Technologie; Ernakulam; On-site; INR 1,800,000 - 2,200,000; Full time

Job summary

Milestone-Technologie is seeking an experienced AI engineer to implement and test capabilities across feature suites, including GraphRAG-based query generation and prompt tuning. The role requires strong Python skills, extensive RAG experience, and familiarity with SPARQL and graph-aware retrieval.

You will collaborate with an AI Architect and Backend Engineer to deliver orchestration integration and support knowledge-graph evolution while expanding the golden dataset for grounding LLM systems.

Qualifications

• 6+ years AI engineering experience with strong Python fundamentals.
• Hands-on RAG pipeline and prompt-engineering experience.
• LLM application development experience.
• Comfortable validating work against a capped, governed golden dataset.
• 3+ years RAG / GraphRAG implementation experience.
• Vector store integration experience.
• Familiarity with SPARQL and graph-aware retrieval.
• 3+ years NLP / semantic engineering experience.
• Ontology mapping and semantic disambiguation experience.
• Experience building or validating golden Q&A datasets for grounding LLM systems.

Responsibilities

• Develop feature logic for assigned capabilities within feature suites.
• Design and tune prompts and RAG pipeline integrations.
• Validate sprint-level acceptance criteria against the golden dataset baseline.
• Coordinate with the AI Architect and Backend Engineer on design and integration questions.
• Build GraphRAG-based dynamic query generation to replace template-based SPARQL querying.
• Deliver graph-aware retrieval agent cards.
• Support knowledge-graph schema evolution.
• Coordinate with the KG validation workstream on schema and validation questions.
• Deliver full orchestration integration with the platform's cookbook-based orchestration component.
• Integrate the RAG-layer retrieval pipeline to inform agent selection, routing and response generation.
• Document the upgrade path to fine-tuning (documentation only, not an implementation, in this phase).
• Expand the golden dataset and semantic grounding.
• Maintain question-and-answer baseline categorization.

Job description

Role summary

Implements and tests assigned capabilities across feature suites, including RAG pipeline integration and prompt tuning; delivers GraphRAG-based dynamic query generation, graph-aware retrieval, and knowledge-graph schema evolution support; delivers full orchestration integration for the platform's orchestration component.

Key responsibilities

• Develop feature logic for assigned capabilities within assigned feature suites
• Design and tune prompts and RAG pipeline integrations
• Validate sprint-level acceptance criteria against the golden dataset baseline
• Coordinate with the AI Architect and Backend Engineer on design and integration questions
• Build GraphRAG-based dynamic query generation to replace template-based SPARQL querying
• Deliver graph-aware retrieval agent cards
• Support knowledge-graph schema evolution
• Coordinate with the KG validation workstream on schema and validation questions
• Deliver full orchestration integration with the platform's cookbook-based orchestration component
• Integrate the RAG-layer retrieval pipeline to inform agent selection, routing and response generation
• Document the upgrade path to fine-tuning (documentation only, not an implementation, in this phase)
• Expand the golden dataset and semantic grounding
• Maintain question-and-answer baseline categorization

Required skills & experience

• 6+ years AI engineering experience with strong Python fundamentals
• Hands-on RAG pipeline and prompt-engineering experience
• LLM application development experience
• Comfortable validating work against a capped, governed golden dataset rather than open-ended tuning
• 3+ years RAG / GraphRAG implementation experience
• Vector store integration experience
• Familiarity with SPARQL and graph-aware retrieval
• Strong Python skills
• 3+ years RAG implementation and orchestration integration experience
• Vector store / retrieval pipeline integration experience
• Comfortable scoping a fine-tuning upgrade path as a design recommendation rather than an implementation
• 3+ years NLP / semantic engineering experience
• Ontology mapping and semantic disambiguation experience
• Experience building or validating golden Q&A datasets for grounding LLM systems

Preferred qualifications

• Experience on a regulated or enterprise AI product
• Familiarity with LangGraph or comparable agent frameworks
• Direct experience replacing a template-based SPARQL approach with GraphRAG
• Experience coordinating with a separate KG validation team
• Experience with cookbook-style orchestration frameworks
• Experience writing forward-looking technical design documents (vs. shipping code) when scope calls for it
• Experience working within a per-sprint expansion cap rather than open-ended dataset growth
• Pharma/manufacturing terminology disambiguation experience

Education

B.Tech/MCA in Computer Science, AI/ML, or a related field, or equivalent practical experience.