Senior Analyst – Agentic AI Engineer with strong hands-on expertise in designing, building, and deploying LLM-powered and Agentic AI applications.

About the Company

LatentView Analytics is a globally recognized data analytics and AI consulting firm headquartered in Princeton, New Jersey, with a strong delivery presence in Chennai, India. With over two decades of industry experience, LatentView partners with Fortune 500 enterprises across retail, CPG, financial services, healthcare, and technology sectors to deliver data-driven business outcomes. The company is known for its deep expertise in advanced analytics, machine learning, and AI engineering, and has built a reputation for helping clients transform raw data into strategic assets. LatentView fosters a culture of innovation, continuous learning, and technical excellence, offering engineers the opportunity to work on cutting-edge AI projects at scale.

Role Overview

We are seeking a highly skilled Senior Analyst – Agentic AI Engineer to join our AI Engineering team in Chennai. This is a hands-on, production-focused role for someone who has deep experience designing, building, and deploying LLM-powered and Agentic AI applications in real-world enterprise environments. You will be responsible for architecting multi-agent systems, developing autonomous workflows, building RAG-based applications, creating robust evaluation frameworks, and deploying AI solutions that are reliable, scalable, and high-quality. The ideal candidate brings 4 to 6 years of professional experience in AI Engineering, Machine Learning, Data Science, or Software Engineering, with a strong track record of taking GenAI solutions from prototype to production. You will work closely with data engineering, ML platform, product, and business teams to deliver next-generation AI automation.

Key Responsibilities

• Design and develop multi-agent architectures involving generation agents, evaluation/critic agents, and orchestration agents to solve complex enterprise use cases.
• Build and deploy scalable LLM-powered applications and generation pipelines that consistently deliver high-quality outputs.
• Develop LLM-as-a-Judge, critic-agent, and automated evaluation frameworks to assess output quality, accuracy, relevance, and compliance.
• Design and implement effective prompt engineering, grounding, Retrieval-Augmented Generation (RAG), and vector search strategies using trusted enterprise data sources.
• Develop and maintain agent orchestration workflows using frameworks such as LangGraph, LangChain, CrewAI, AutoGen, or similar technologies.
• Implement human-in-the-loop approval workflows and feedback mechanisms to continuously improve agent performance and reliability.
• Build and implement AI guardrails, validation layers, and content governance mechanisms for production-grade AI applications.
• Integrate applications with major LLM providers and APIs, and evaluate models based on quality, latency, scalability, and cost.
• Deploy and support LLM and Agentic AI applications on cloud AI/ML platforms such as Databricks Mosaic AI, Model Serving, or equivalent platforms.
• Implement LLMOps and MLOps practices, including experiment tracking, prompt/version management, monitoring, tracing, evaluation, and observability.
• Monitor and optimize AI applications for latency, performance, scalability, token usage, cost, and output quality.
• Collaborate with cross-functional teams (data engineering, machine learning, platform, product, and business) to develop and deliver scalable Agentic AI solutions.

Tech Stack

• Programming Languages: Python (advanced, production-grade)
• Agentic AI Frameworks: LangGraph, LangChain, CrewAI, AutoGen, or similar
• LLM & GenAI: Large Language Models (GPT, Claude, open-source models), RAG, vector search, prompt engineering, LLM evaluation techniques (LLM-as-a-Judge, critic agents)
• Vector Databases: Pinecone, Weaviate, Chroma, or similar
• Cloud AI/ML Platforms: Databricks Mosaic AI, Model Serving, or equivalent (AWS SageMaker, GCP Vertex AI)
• MLOps/LLMOps: Experiment tracking (MLflow, Weights & Biases), prompt management, version control, monitoring, tracing, observability
• AI Guardrails & Governance: Responsible AI frameworks, validation layers, content safety tools
• APIs & Integration: RESTful APIs, LLM provider APIs (OpenAI, Anthropic, etc.)
• Other: Human-in-the-loop workflows, cost optimization techniques, scalable architecture patterns

4 to 6 years of professional experience in AI Engineering, Machine Learning Engineering, Data Science, Software Engineering, or related technical roles. Strong hands-on experience with Python and developing production-grade applications. Experience building and deploying LLM-powered or Generative AI applications. Hands-on experience designing and implementing Agentic AI and multi-agent systems. Experience with Agentic AI and orchestration frameworks such as LangGraph, LangChain, CrewAI, AutoGen, or similar technologies. Strong knowledge of Prompt Engineering, Retrieval-Augmented Generation (RAG), grounding techniques, vector databases, and semantic search. Experience working with major LLM providers and APIs and understanding model selection trade-offs related to cost, latency, performance, and output quality. Hands-on experience with LLM evaluation techniques, including LLM-as-a-Judge, critic agents, automated evaluation, and quality assessment frameworks. Understanding of AI guardrails, validation frameworks, Responsible AI, content governance, and safety controls. Experience implementing or working with human-in-the-loop workflows and feedback mechanisms. Experience deploying LLM or Agentic AI applications on cloud AI/ML platforms such as Databricks Mosaic AI, Model Serving, or equivalent platforms. Knowledge of MLOps / LLMOps practices, including experiment tracking, prompt management, version control, monitoring, tracing, and observability would be an added advantage. Understanding of AI application optimization related to latency, scalability, token consumption, and cost management. Experience taking AI/LLM applications from proof-of-concept or prototype stages to production environments. Strong problem-solving skills and the ability to work effectively with cross-functional engineering and business teams.