Software Engineer – Data Science

What You'll Own

• Spotting where AI and ML fit: product features and business problems that machine learning can solve.
• ML algorithms and solutions in Python, from design through implementation.
• Agentic AI applications built on LangChain, LangGraph, or similar frameworks.
• The path from problem definition and PoC to production deployment, often on early-stage products where the problem is still being shaped.
• Integration of your AI/ML solutions into production systems, working with engineering managers and product engineering teams.
• AI features that ship inside startup and product-company client's products, built around the business problems those clients need solved.
• Model quality after launch: performance evaluation, algorithm optimization, and the reliability, scalability, and maintainability of what you deploy.
• Applying research papers, emerging techniques, and industry best practices to practical problems.
• Your share of the Data Science team's technical discussions and knowledge sharing.

What We Need

• A strong foundation in Machine Learning, with a background primarily in ML.
• At least 1 year of hands-on experience in Agentic AI, including the development of AI agents or agent-based applications.
• Production experience in developing and deploying AI/ML solutions.
• Hands-on experience with LangChain, LangGraph, or similar frameworks for building LLM- powered applications or AI agents.
• Strong programming skills in Python and the ability to implement algorithms and develop maintainable code.
• Good understanding of ML algorithms, statistical techniques, and model evaluation methods.
• Practical experience in at least one area of Machine Learning, such as NLP, Computer Vision, or Predictive Analytics.
• Experience with standard ML/DL libraries such as Scikit-learn, TensorFlow, Keras, or OpenCV, as relevant to the problem domain.
• Ability to convert PoCs into production-quality solutions and integrate them into existing applications.
• Good analytical, problem-solving, communication, and collaboration skills.

Good to have

• Experience with Large Language Models (LLMs), prompt engineering, Retrieval-Augmented Generation (RAG), and vector databases.
• Familiarity with agent orchestration, tool calling, multi-step reasoning workflows, and evaluation of AI agent outputs.
• Experience with cloud platforms, APIs, Docker, or MLOps practices.
• Exposure to Big Data technologies.
• Ability to understand and apply relevant research papers to practical ML/AI problems.
• Relevant coursework or certifications in Machine Learning, statistics, linear algebra, or deep learning.

Must have

• Machine Learning, Python, Agentic AI (1+ year), LangChain/LangGraph or similar frameworks, and production deployment experience.

Why It's Worth Your Time

• You'll build with LLMs, AI agents, and agent orchestration as your core daily work, bringing new techniques from research into products as they mature.
• The team has deployed 25+ ML solutions across predictive analytics, time-series forecasting, NLP, virtual assistants, and object detection, and working across problems that varied is how you'll build range as an ML engineer.
• You'll take products for US-based startups from first prototype to production, which builds speed and breadth faster than working inside one mature system.
• You'll be guided by hands-on data science leads who review your model choices, evaluation approach, and agent designs, and help you take PoCs to production.
• You'll be surrounded by talented peers with strong educational backgrounds, and pushed to grow into a stronger AI systems designer alongside them.

About Talentica

Since 2003, Talentica has built software exclusively for tech companies — from early-stage startups to enterprises. We've partnered with startups from their earliest stages through to successful acquisitions, doing the kind of deep product and engineering work that moves a company's valuation, not generalist consulting or staff augmentation. Our ~600-person team is among the strongest in the industry, drawn heavily from the IITs and NITs — which means the solutions you'll build are genuinely differentiated, and the work is demanding enough that you'll keep learning regardless of how senior you are.