Website:
advocateatulsharma.com
Job details:
Role Description
This is a full-time, on-site role based in Delhi, India, for an Agentic AI/ML & Generative AI Architect. The architect will design, build, and maintain AI and agentic workflows and Day-to-day responsibilities include translating legal use cases into AI system architectures, integrating large language models with legal databases, and implementing tools for vector search, entity extraction, and risk assessment. The role involves collaborating closely with the advocate to automate repetitive tasks, ensure data security and compliance with legal-ethical standards, and continuously improve model performance based on real-world case outcomes. The architect will also monitor production systems, troubleshoot issues, document solutions, and train team members in the effective and responsible use of AI tools.
Qualifications
- Strong foundation in machine learning and deep learning, including experience with supervised/unsupervised learning, model training, evaluation, and deployment.
- Hands-on experience with Generative AI and large language models (e.g., GPT, LLaMA, or similar), including prompt engineering, fine-tuning, and retrieval-augmented generation (RAG).
- Proficiency in Python and common AI/ML frameworks (such as PyTorch, TensorFlow, or JAX), as well as NLP libraries (e.g., Hugging Face, spaCy, or similar).
- Experience architecting agentic AI systems and workflows (multi-agent orchestration, tools integration, connectors, and pipelines) to automate complex, multi-step tasks.
- Knowledge of data engineering practices, including data pipelines, ETL, feature engineering, and working with structured and unstructured text data at scale.
- Understanding of information retrieval, vector databases, semantic search, and document intelligence solutions relevant to legal or compliance-heavy domains.
- Familiarity with software engineering best practices (version control
Click on Apply to know more.