About this role
Join Prismforce’s AI & ML team in Bangalore to build NLP and machine-learning solutions for talent intelligence. The posting covers Data Scientist I and Data Scientist II; its overall experience heading states 1–4 years. The role spans data exploration, model development, evaluation and deployment. It is full-time with a hybrid work model.
Responsibilities
- Develop and fine-tune models for named entity recognition, text classification, semantic similarity and clustering. Use pretrained embeddings, transformers and LLMs to extract and normalize structured data from messy text.
- Design experiments, evaluate performance and work with product and engineering teams to make models robust and production-ready. Explore research and open-source tools to improve scalability.
- At the Data Scientist II level, own modules from problem scoping through deployment and monitoring; productionize solutions for ambiguous problems; design evaluation frameworks with baselines, benchmarks, and offline and live metrics; review code and mentor junior data scientists.
Requirements
- For Data Scientist I, the detailed requirements state 1–3 years of hands-on applied NLP and ML experience, strong Python skills, experience with Pandas, Scikit-learn, and PyTorch or TensorFlow, and familiarity with spaCy, Hugging Face Transformers and SBERT. Understand embedding models, attention mechanisms and evaluation metrics, and be able to develop data-driven solutions to abstract problems.
- For Data Scientist II, the detailed requirements state 3–5 years in applied NLP/ML and at least one productized solution. Understand modeling tradeoffs, labeling strategies and error analysis; have experience customizing transformers and familiarity with pipeline orchestration, versioning and ML lifecycle practices.
Preferred experience includes LLM APIs, prompt engineering, LLM-based metadata generation, knowledge graphs, taxonomy design, vector databases, open-source NLP contributions or research. Additional Data Scientist II bonuses include embedding alignment, taxonomy bootstrapping, evaluation at scale, drift detection, human-in-the-loop systems, vector search and LLM agent orchestration. The posting also lists LoRA, QLoRA, LLM training and fine-tuning as required skills. Benefits mentioned include insurance coverage, retirement benefits, flexible policies and career-development opportunities.