Large Language Model Architect

Overview

Architect large language models that process and generate natural language, designing neural network parameters trained on large quantities of unlabeled text data. The role also assesses whether proposed Generative AI and Agentic AI initiatives are technically viable before full-scale implementation.

What you'll do

• Architect large language models (LLMs) that can process and generate natural language.
• Design neural network parameters trained on large quantities of unlabeled text data.
• Conduct proof-of-concepts (PoCs) and pilot implementations.
• Perform architecture validation and integration assessments.
• Evaluate GenAI and Agentic AI use cases for technical viability and implementation risk.
• Assess suitability of large language models such as OpenAI GPT models, Anthropic Claude, and Google DeepMind Gemini for business use cases.
• Evaluate Retrieval-Augmented Generation (RAG), fine-tuning, and prompt engineering strategies.
• Validate model performance using structured evaluation metrics.
• Prototype agent orchestration using frameworks such as LangChain, AutoGPT, and Microsoft AutoGen.
• Define guardrails, human-in-the-loop mechanisms, and observability requirements.
• Conduct adversarial testing and bias evaluation.
• Perform cost modeling and ROI analysis for AI initiatives.
• Identify technical, operational, and security risks in proposed AI solutions.
• Translate complex AI feasibility findings into executive-ready insights.
• Provide go/no-go recommendations based on structured feasibility frameworks.
• Partner with Architecture, Engineering, Security, and Product teams.

What you'll need

• Must have skills: Large Language Models (LLMs).
• Minimum 12 years of experience is required.
• 15 years of full-time education.
• Familiarity with vector databases and embedding pipelines.
• Experience in cost modeling and ROI analysis for AI initiatives.
• Experience evaluating AI platforms across Amazon Web Services, Microsoft Azure, and Google Cloud.
• Deep understanding of LLM architectures and transformers, RAG pipelines, multi-agent systems, cloud-native architecture, and MLOps / LLMOps practices.

Details

• Location: Hyderabad.