Agentic AI Engineer

Client: TCS | Engagement: Full-time | Work mode: HYBRID | Experience: 8+ | Publisher job id: JOB-000656

Role Overview

We are seeking an experienced AI/ML & Forward Deployed Engineer with 8+ years of engineering expertise to deliver high-impact AI/ML and GenAI solutions end-to-end. This role involves blending applied machine learning, software engineering, and stakeholder problem-solving to deploy production-grade systems that are scalable, secure, observable, and aligned to business KPIs. The Agentic AI Engineer is a hands-on development role at TCS (Americas) specializing in building and deploying AI agent solutions for clients. This client-facing consulting position operates in a hybrid environment, focusing on delivering cutting-edge AI agents that integrate large language models, custom prompts, data sources, and business logic. Projects can range from financial chatbots to manufacturing optimizers, requiring advanced prompt engineering, Retrieval-Augmented Generation (RAG), and strong software skills. Key Responsibilities: Partner with stakeholders (business/product/customers) to identify and shape AI opportunities into well-defined use cases with success metrics, constraints, and rollout plans. Run workshops and technical discovery to assess feasibility, data readiness, integration needs, and operational risks. Drive rapid prototyping, pilot deployments, and iterative improvements based on real user feedback. Develop and improve ML solutions including classification, regression, ranking, forecasting, anomaly detection, and NLP. Establish and maintain robust evaluation practices: offline metrics, validation strategies, experimentation, and A/B testing. Perform feature engineering, error analysis, model optimization, and performance tuning for production requirements. Build and productionize RAG (Retrieval-Augmented Generation) pipelines, including document ingestion, chunking strategy, embeddings, retrieval tuning, reranking, and response grounding. Implement guardrails and reliability patterns: prompt templates, tool/function calling, hallucination reduction, citation strategies, and fallback paths. Code the core logic for AI agents, whether standalone or in multi-agent systems, enabling them to answer questions, generate content, or execute transactions. Use Python or similar languages to integrate large language models (LLMs) and external tools (e.g., APIs, web search, databases). Craft, refine, and test prompts to guide agent behavior, including fallback strategies for uncertainty. Connect AI agents to vector databases or search indices to ground outputs in up-to-date, domain-specific information. Integrate AI agents with external systems (e.g., travel booking APIs, payment gateways), handling formatting, RESTful calls, and data responses as needed. Simulate agent behavior, identify and fix failure modes, and tune prompts and code for high-quality results. Package and deploy agent applications (Docker, cloud), ensuring scalability and proper configuration. Work with AI Architects, Data Engineers, and UX Developers in agile teams, contributing to sprints and client demos. Tailor solutions for each industry, adapting compliance, personalization, and integration as needed. Implement guardrails, content moderation, and privacy measures, following TCS’s responsible AI guidelines. Required Skills: Expertise in Python (and optionally Java, JavaScript, or C), unit testing, and version control (Git). Solid grasp of machine learning and AI concepts, model behavior, and experience with NLP or chatbots. Experience crafting and iterating prompts, including few-shot examples and output formatting techniques. Familiarity with embedding models, vector databases, and unstructured data processing. Building and consuming RESTful APIs, microservices, and handling JSON/XML data formats. Knowledge of lists, dictionaries, trees/graphs, and their application in efficient agent design. Mastery of Python for AI/ML, with exposure to JavaScript/TypeScript, FastAPI, or Flask for APIs. Experience with AI model APIs (OpenAI, Azure OpenAI), and ML frameworks like PyTorch or TensorFlow. Hands-on with LangChain or similar frameworks for prompt management and agent logic. Working with SQL, NoSQL, and vector databases (e.g., Pinecone, Weaviate) for data retrieval. Familiarity with Docker, CI/CD, and cloud deployment (AWS, Azure, GCP, Lambda/Functions). Proficient with Git and DevOps platforms (GitHub, GitLab, Bitbucket). Experience with PyTest, Postman, and AI evaluation methods. Practical knowledge of cloud AI offerings and environment configuration. Implementing logging (Python logging, CloudWatch). Qualifications: 8+ years of engineering experience. Preferred Skills: Experience with event-driven workflows (RabbitMQ, Kafka, SQS).