Data Science Manager

Job Description:

Graduate degree in a quantitative field (CS, statistics, applied mathematics, machine learning, or related discipline) • Good programming skills in Python with strong working knowledge of Pythons numerical, data analysis, or AI frameworks such as NumPy, Pandas, Scikit-learn, etc. • Experience with LMs (Llama (1/2/3), T5, Falcon, Langchain or framework similar like Langchain) • Candidate must be aware of entire evolution history of NLP (Traditional Language Models to Modern Large Language Models), training data creation, training set-up and finetuning • Candidate must be comfortable interpreting research papers and architecture diagrams of Language Models • Candidate must be comfortable with LORA, RAG, Instruct fine-tuning, Quantization, etc. • Predictive modelling experience in Python (Time Series/ Multivariable/ Causal) • Experience applying various machine learning techniques and understanding the key parameters that affect their performance • Experience of building systems that capture and utilize large data sets to quantify performance via metrics or KPIs • Excellent verbal and written communication • Comfortable working in a dynamic, fast-paced, innovative environment with several ongoing concurrent projects.

Roles & Responsibilities:

• Lead a team of Data Engineers, Analysts and Data scientists to carry out following activities: • Connect with internal / external POC to understand the business requirements • Coordinate with right POC to gather all relevant data artifacts, anecdotes, and hypothesis • Create project plan and sprints for milestones / deliverables • Spin VM, create and optimize clusters for Data Science workflows • Create data pipelines to ingest data effectively • Assure the quality of data with proactive checks and resolve the gaps • Carry out EDA, Feature Engineering & Define performance metrics prior to run relevant ML/DL algorithms • Research whether similar solutions have been already developed before building ML models • Create optimized data models to query relevant data efficiently • Run relevant ML / DL algorithms for business goal seek • Optimize and validate these ML / DL models to scale • Create light applications, simulators, and scenario builders to help business consume the end outputs • Create test cases and test the codes pre-production for possible bugs and resolve these bugs proactively • Integrate and operationalize the models in client ecosystem • Document project artifacts and log failures and exceptions. • Measure, a