About the role
Responsibilities:
- Analyze product requirements and come up with data science solutions.
- Identify and develop data engineering scripts (example: parsers) necessary to build training datasets.
- Build deep learning NLP model(s), customize as needed to meet the requirements.
- Write production-quality Python code for model development as well as for inference.
- Ability to think out-of-the-box and implement custom loss functions and quantitative methods to increase the accuracy of AI solutions.
- Build and ship AI agent-driven systems that reason, plan, and act across real-world, messy data.
- Design agentic workflows using LLMs, tools, retrieval, and memory to solve high-impact product problems.
- Work hands-on with unstructured and multimodal data (text, PDFs, drawings, images, logs).
- Develop multimodal models (text + vision) for document understanding and contextual reasoning.
- Own AI solutions end-to-end: data pipelines, modeling, deployment, monitoring, and iteration.
- Write production-grade Python code and rapidly prototype, test, and ship in a cloud-native environment.
- Take complete ownership of the solution in all phases: analysis, proof of concept, data engineering, model development, model tuning, and model implementation.
Requirements:
- 7+ years of experience building production ML / AI systems, with recent experience in LLMs or AI agents in data science.
- Strong Python engineer with a bias toward clean, scalable, and maintainable code.
- Proven experience working with unstructured data and images at scale.
- Hands-on experience with agent patterns (tool use, function calling, planning, memory).
- Good proficiency and hands-on experience with LLMs and advanced AI models
- Experience deploying AI systems on AWS, GCP, or Azure in microservices architectures.
- Thrives in fast-moving startup environments with high ownership and ambiguity.
- Strong Python skills with a focus on data engineering and data analysis.
- Proficiency with data mining algorithms such as Scikit-Learn, NumPy, SciPy, and Pandas.
- Strong understanding of machine learning models, model training, and hyperparameter tuning.
- Working knowledge of deep learning models, loss functions, and accuracy measures.
- Hands-on proficiency with PyTorch / TensorFlow / Keras.
- Experience in NLP solutions preferred.
- Experience in parsing PDF and text documents preferred.
- Experience with statistical regression, neural nets, deep learning, decision trees, SVM, and ensembles is expected.
- Multi-cloud experience and proficiency with providers AWS, GCP, or Azure
- Comfortable working in a microservices environment
- Self-motivated, enthusiastic to build next-generation AI systems.