Key Responsibilities

- Build and maintain RAG pipelines, prompt/agent workflows, and evaluation harnesses for internal engineering use cases
- Support fine-tuning and domain adaptation of LLMs for automotive requirements engineering and compliance documentation
- Work on reducing hallucination and improving reliability via knowledge-graph grounding and auto-evaluation frameworks
- Prototype AI-assisted tools for requirements drafting, traceability, and test-case generation supporting the ASPICE pipeline
- Explore applied-AI use cases beyond engineering (business development intelligence, HR process automation) under the 'AI+X' mandate
- Document and present findings and working prototypes to engineering and product stakeholders

A Typical Day

- Morning: check evaluation dashboards and overnight pipeline runs; triage failures and hallucination/quality regressions
- Morning: build and iterate on RAG pipelines, prompts, and agent workflows for the ASPICE SOW-to-artifact use case
- Mid-day: pair with a Systems Software Engineer to test generated requirements or test cases against real project data and capture feedback
- Afternoon: dataset preparation, fine-tuning or domain-adaptation experiments, knowledge-graph grounding work
- Afternoon:



extend the evaluation harness — add cases, measure accuracy and hallucination rate, log results
- Late day: document findings, update the prototype backlog, prepare short demos for stakeholders
- Weekly rhythm: review with the GenAI/Applied AI Lead on use-case prioritisation; periodic 'AI+X' discovery sessions with SW, HW, BD and HR

Required Skills

- Strong foundation in Python and NLP/deep learning fundamentals
- Familiarity with LLM APIs (Anthropic, OpenAI, or similar) and RAG concepts
- Understanding of prompt engineering and basic agent-framework concepts
- Analytical mindset and comfort with ambiguity typical of a proving-ground role

Positive-to-Have Skills

- Exposure to automotive or engineering-domain data
- Experience with workflow-automation frameworks (e.g. n8n)
- Familiarity with evaluation/observability practices for LLM systems (hallucination reduction, auto-eval)

Standards & Compliance Exposure No domain-standard compliance required at this level; awareness of ASPICE terminology is a plus given the pipeline it supports

Education

Bachelor's/Master's in Computer Science, AI/ML, or related field

Apply Now

📌 GenAI/Applied AI (Pune)
🏢 Interface Microsystems
📍 Pune