Data Scientist

Data Scientist

Responsibilities

●Design and implement AI agent workflows. Develop end-to-end intelligent pipelines and multi-agent systems (e.g., LangGraph/LangChain workflows) that coordinate multiple LLM-powered agents to solve complex tasks. Create graph-based or state-machine architectures for AI agents, chaining prompts and tools as needed.

●Build and fine-tune generative models. Develop, train, and fine-tune advanced generative models (transformers, diffusion models, VAEs, GANs, etc.) on domain-specific data. Deploy and optimize foundation models (such as GPT, LLaMA, Mistral) in production, adapting them to our use cases through prompt engineering and supervised fine-tuning.

●Develop data pipelines. Build robust data collection, preprocessing, and synthetic data generation pipelines to feed training and inference workflows. Implement data cleansing, annotation, and augmentation processes to ensure high-quality inputs for model training and evaluation.

●Implement LLM-based agents and automation. Integrate generative AI agents (e.g., chatbots, AI copilots, content generators) into business processes to automate data processing and decision-making tasks. Use Retrieval-Augmented Generation (RAG) pipelines and external knowledge sources to enhance agent capabilities. Leverage multimodal inputs when applicable.

●Optimize performance and safety. Continuously evaluate and improve model/system performance. Use GenAI-specific benchmarks and metrics (e.g., BLEU, ROUGE, TruthfulQA) to assess results, and iterate to optimize accuracy, latency, and resource efficiency. Implement safeguards and monitoring to mitigate issues like bias, hallucination, or inappropriate outputs.

●Collaborate and document. Work closely with product managers, engineers, and other stakeholders to gather requirements and integrate AI solutions into production systems. Document data workflows, model architectures, and experimentation results. Maintain code and tooling (prompt libraries, model registries) to ensure reproducibility and knowledge sharing.

Required Skills & Qualifications

●Education: Bachelor’s or Master’s degree in Computer Science, Data Science, Artificial Intelligence, or a related quantitative fieldanalyticsvidhya.com (or equivalent practical experience). A strong foundation in algorithms, statistics, and software engineering is expected.

●Programming proficiency: Expert-level skills in Pythoncoursera.org, with hands-on experience in machine learning and deep learning frameworks (PyTorch, TensorFlow)analyticsvidhya.com. Comfortable writing production-quality code and using version control, testing, and code review workflows.

●Generative model expertise: Demonstrated ability to build, fine-tune, and deploy large-scale generative modelsanalyticsvidhya.com. Familiarity with transformer architectures and generative techniques (LLMs, diffusion models, GANs)analyticsvidhya.comanalyticsvidhya.com. Experience working with model repositories and fine-tuning frameworks (Hugging Face, etc.).

●LLM and agent frameworks: Strong understanding of LLM-based systems and agent-oriented AI patterns. Experience with frameworks like LangGraph/LangChain or similar multi-agent platformsgyliu513.medium.com. Knowledge of agent communication standards (e.g., MCP/Agent Protocol)gyliu513.medium.comblog.langchain.dev to enable interoperability between AI agents.

●AI integration and MLOps: Experience integrating AI components with existing systems via APIs and services. Proficiency in retrieval-augmented generation (RAG) setups, vector databases, and prompt engineeringanalyticsvidhya.com. Familiarity with machine learning deployment and MLOps tools (Docker, Kubernetes, MLflow, KServe, etc.) for managing end-to-end automation and scalable workflowsanalyticsvidhya.com.

●Familiarity with GenAI tools: Hands-on experience with state-of-the-art GenAI models and APIs (OpenAI GPT, Anthropic, Claude, etc.) and with popular libraries (Hugging Face Transformers, LangChain, etc.). Awareness of the current GenAI tooling ecosystem and best practices.

●Soft skills: Excellent problem-solving and analytical abilities. Strong communication and teamwork skills to collaborate across data, engineering, and business teams. Attention to detail and a quality-oriented mindset. (See Ideal Candidate below for more on personal attributes.)

Ideal Candidate

●Innovative, problem-solver: You are a creative thinker who enjoys tackling open-ended challenges. You have a solutions-oriented mindset and proactively experiment with new ideas and techniquesanalyticsvidhya.com.

●Systems thinker: You understand how different components (data, models, services) fit together in a large system. You can architect end-to-end AI solutions with attention to reliability, scalability, and integration points.

●Collaborative communicator: You work effectively in multidisciplinary teams. You are able to explain complex technical concepts to non-technical stakeholders and incorporate feedback. You value knowledge sharing and mentorship.

●Adaptable learner: The generative AI landscape evolves rapidly. You are passionate about staying current with the latest research and tools. You embrace continuous learning and are eager to upskill and try new libraries or platformsanalyticsvidhya.com.

●Ethical and conscientious: You care about the real-world impact of AI systems. You take responsibility for the quality and fairness of models, and proactively address concerns like data privacy, bias, and security