About the role
Responsibilities:
- Fine-tuning pipelines with LoRA / QLoRA / PEFT on domain-specific healthcare corpora.
- Synthetic training data generation (JSONL) for clinical and billing NLP tasks.
- Evaluation design: precision, recall and task-specific rubrics for healthcare language.
- Quantisation and inference optimisation (GGUF, AWQ, bitsandbytes) for GPU-constrained deployments.
- Systematic, reproducible prompt engineering.
- Benchmarking SLMs against frontier models on narrow, well-defined tasks.
Requirements:
- M. Tech in AI/ML from IISc
- 5-6 years of experience, with hands-on LLM/SLM fine-tuning.
- Experience shipping models in a product company, ideally healthtech or another regulated domain.
- Strong evaluation instincts: you measure before you claim.