ML Engineer (Classical ML + GenAI)

ML Engineer (Classical ML + GenAI)

Location - Hyderabad

Experience - 4+ years

This role delivers AI use cases from first principles to production. The profile spans classical ML (predictive modelling, process optimisation) and GenAI (LLM-powered applications, RAG, agents) — not a specialist in one, but genuinely capable across both. The defining quality is the ability to take a business problem, select the right technical approach, and see it through to a reliable, deployed product — not just a proof of concept.

CORE RESPONSIBILITIES

– Own the full technical lifecycle of AI use cases: problem and mvp scoping → data analysis → model/application development → pilot → productionisation

– Build GenAI applications: RAG pipelines, LLM-powered features, agents, and prompt orchestration workflows for classical and more manufacturing related use cases

– Build and productionise classical ML and Deep Learning models for manufacturing use cases (e.g., predictive maintenance, smart allocation, predictive DFM )

– Evaluate and iterate — define success metrics, run experiments, measure model and application performance in production

KEY SKILLS

– Classical ML and Deep Learning experience

– GenAI: LLM APIs,RAG patterns, LangChain / LlamaIndex, fine-tuning, prompt engineering at scale – Data: can prepare their own datasets, experience in data processing for structured and highly unstructured data

– Productionisation: writing clean, testable code; working with Docker; understanding how their models will be served

– Evaluation mindset: knows how to define and measure quality for both ML models and GenAI applications

WHAT GOOD LOOKS LIKE

– Has taken at least one classical ML use case, one Deep Learning and one complex GenAI use case from prototype to production

– Can write production-quality code and understands what it takes to deploy reliably

– Comfortable with ambiguity in problem definition — can scope progressive MVP scopes to allow early value

– Good engineering instincts: doesn't over-engineer, but doesn't produce fragile one-off scripts either

Would be great if the candidate has some experience in applying AI to complex engineering data (e.g., 3D geometries, complex documents)

WHAT THIS ROLE IS NOT

– Not a pure research scientist — the bar is production delivery, not publication – Not an LLM specialist only

— classical ML and DL use cases are equally in scope and require genuine capability on diverse data and use cases