NLP Engineer
Jobgether (company site, opens in a new tab)RemoteDirect career pageEst. 14–28 LPA
Posted 5d ago
Reach people at Jobgether
People at this company you can reach
Log in to see who works here
Create a free account or log in to see the people at Jobgether you could write to:
- Recruiters & HR
- People in this function
- Work addresses
Your match
Against your primary resume
Log in to see your match score
Create a free account or log in to unlock your CV match score across:
- Role fit
- Experience
- Industry relevance
Protocol Intelligence
How competitive this role is
Competition
Background fit
Strictness
Log in to see why each signal reads the way it does.
Job description
Structured overview of role & requirements
About This Role
- Own the language-model lifecycle for conversational assistants and copilots: prepare datasets, fine-tune models, evaluate quality, and improve responses.
- Build and operate production model APIs; optimize inference speed, GPU use, latency, and serving costs, and handle monitoring, failures, and rollbacks.
- Compare models and prompting approaches, using benchmarks and user feedback to balance quality, reliability, and cost; contribute to recommendation capabilities.
Minimum Requirements
- Work Experience Required: Not explicitly mentioned in the JD.
- Strong Python plus practical PyTorch and Hugging Face Transformers experience; hands-on SFT and parameter-efficient fine-tuning such as LoRA are required.
- Experience with dataset quality and leakage prevention, generated-text evaluation, model APIs and prompt engineering, and production APIs using FastAPI or an equivalent framework.
- Familiarity with Docker, Linux, Git, GPU deployment, and inference/serving frameworks such as vLLM or equivalent; SQL proficiency and practical recommendation-systems experience (including project-level) are required. The role is based in India and fully remote.
Ideal Candidate Profile
- Has built beyond research notebooks, taking language models through data preparation, evaluation, deployment, and production support.
- Can diagnose model-quality issues by connecting evaluation results and user feedback to dataset gaps, training changes, or prompt/model choices.
- Fits a production-focused, distributed engineering environment where model quality must be balanced against latency, reliability, and operating cost.