1 day ago
Pune, India or Bengaluru, IndiaSenior
Responsibilities
- Design, develop, maintain, test, and deploy LLM-powered generative AI applications.
- Build and orchestrate multi-agent systems with inter-agent communication, task routing, and tool orchestration.
- Implement RAG workflows including embedding creation, chunking, vector indexing, and context-aware retrieval.
- Apply prompt-engineering techniques to improve model reliability and interpretability.
- Create automated and manual testing pipelines to evaluate model quality, accuracy, safety, and factual grounding.
- Track metrics such as coherence, hallucination rate, and response diversity.
- Implement logging, tracing, token management, monitoring, and optimization for GenAI applications.
- Perform exploratory data analysis and textual data preprocessing for training, evaluation, and fine-tuning.
- Produce design documentation and experiment logs while collaborating with architects, data scientists, UI/UX engineers, and product managers.
- Stay current with new LLM releases and generative AI and agentic AI frameworks.
Requirements
- At least 5 years of total IT experience.
- 2–3 years of experience in AI/ML or generative AI development.
- Hands-on experience building, testing, and deploying LLM-based or agentic AI applications in production or pilot settings.
- Strong proficiency in Python, LangChain, LlamaIndex, CrewAI, and LangGraph.
- Deep understanding of LLM architectures, tokenization, embedding models, and vector databases.
- Experience with prompt testing and evaluation frameworks such as TruLens, LangSmith, PromptLayer, or DeepEval.
- Familiarity with bias mitigation, safety testing, and Responsible AI principles.
- Knowledge of MLOps and LLMOps practices for deploying and monitoring AI models.
- Exposure to at least one major cloud AI platform: Azure, AWS, GCP, or OCI.
- Prior experience with chatbots, intelligent assistants, summarizers, or document automation is preferred.
- Familiarity with open-weight models such as LLaMA, Mistral, or Falcon is preferred.
- Exposure to multimodal AI and enterprise data-source integrations is preferred.
- Experience supporting GenAI projects through testing, benchmarking, and performance optimization is preferred.
Tech Stack
Categories
About PwC
PwC is a global network of professional services firms providing assurance, tax, and advisory work for businesses and public-sector clients. It sells project-based and managed consulting, audit, and deals services, and implements and integrates business applications. Formed in 1998 by the merger of Price Waterhouse and Coopers & Lybrand, it is headquartered in London and is one of the Big Four accounting firms.