Leads the team this role joins
Heads the group this role sits in
In-house, owns the opening
From your own Gmail and LinkedIn.
About the role
This is a full-time, onsite role in Nanakramguda, Hyderabad, India, for an IT developer who builds and operates applied large language model systems. It suits engineers with at least 3 years of software or ML experience, including 2 years of hands-on generative AI or LLM work. The role sits within Medtronic, a healthcare technology company, and covers training, evaluating, serving and governing enterprise LLMs.
What you’ll do
- Design and run training, fine-tuning and adaptation pipelines for open-weight LLMs using SFT, LoRA, QLoRA and other PEFT methods.
- Build and curate datasets from operational sources such as incidents, runbooks, logs, knowledge articles and validated interactions.
- Develop evaluation frameworks covering reasoning, instruction following, hallucination, tool calling, safety, robustness and structured output.
- Improve retrieval-augmented generation through retrieval analysis, embedding comparison, reranking, chunking and groundedness checks.
- Design agent and tool-calling benchmarks that measure task completion, tool selection and failure recovery.
- Optimise serving performance through quantisation, batching, caching and inference tuning.
- Set up experiment tracking, model governance, lineage, release gating and deployment workflows.
- Establish production monitoring, drift detection, security testing and rollback criteria for deployed AI systems.
What they’re looking for
- A bachelor's degree in engineering, an MCA or an MSc, with 3+ years of IT experience.
- 3+ years of software engineering or machine learning experience, including 2+ years of hands-on generative AI or LLM development.
- Experience with Python, PyTorch, Hugging Face Transformers, Datasets, PEFT and fine-tuning workflows.
- Experience with LLM evaluation, retrieval-augmented generation, model benchmarking and experiment tracking frameworks.
- Experience with Linux, Docker, Kubernetes, Git, REST APIs and cloud or GPU-based model deployment.
- Experience with MLflow or Weights & Biases for experiment tracking, model lifecycle management and artifact versioning.
Nice to have
- Experience with direct preference optimisation (DPO), TRL, DeepSpeed, FSDP or other distributed training techniques.
- Experience deploying and optimising models with vLLM, TensorRT-LLM, SGLang or similar serving platforms.
- Experience with RAGAS, DeepEval, TruLens or similar evaluation and observability frameworks.
Skills
Questions about this role
Where is this role based?
The role is onsite in Nanakramguda, Hyderabad, India.
How much experience is needed?
At least 3 years of IT and software or ML experience, including 2 years of hands-on generative AI or LLM work.
Which degree is required?
A bachelor's degree in engineering, an MCA or an MSc.
Is the role full-time?
Yes, it is a full-time position.