Generative AI Platform Engineer
- Tata Consultancy
- 4 - 8 years
- Hyderabad
- 20 days ago
- Email to a friend
- Report this job
Job Description
Dear Candidates,
Greetings from TCS!!!!
TCS is looking for Generative AI Platform Engineer
Experience: 4-8 years
Location: Hyderabad
Date of interview: 6 June-2026
Mode of interview: Virtual
MUST HAVE SKILL:
- LLM Ops & Agent Orchestration: Experience with prompt engineering, prompt/version management, model routing, and evaluation using agentic and orchestration frameworks such as LangChain, LangGraph, Google Agent Development Kit (ADK), and other agent-based LLM frameworks.
- RAG Expertise: Practical experience building and tuning RAG pipelines (chunking strategies, embeddings, retrievers, reranking).
- Vector DBs: Hands-on with at least one: FAISS, Pinecone, Milvus, Weaviate (index types, parameters, scaling).
- Python Engineering: Strong Python with FastAPI/Flask, async IO, typing, packaging; clean code & SOLID principles.
- MLOps/Platform: Docker, Kubernetes, Git, CI/CD (Jenkins/GitHub Actions/Azure DevOps), observability (logs/metrics/traces).
- APIs: Design/consume REST APIs, OpenAPI/Swagger, pagination, error models, rate limiting.
- Data Handling: Experience with text preprocessing, embeddings, metadata schemas, and storage (object stores/RDBMS).
- Build production-grade RAG pipelines, manage vector databases, and own LLM Opsincluding observability, evaluation, and cost/performance optimizationon secure, compliant platforms.
GOOD TO HAVE SKILLS:
- Agentic Architectures: Practical experience comparing and implementing LangChain/LangGraph vs Google ADK and other agentic solutions, understanding tradeoffs across control flow, memory, observability, toolcalling, scalability, and production readiness.
- Retrieval & Ranking: BM25, hybrid search, approximate nearest neighbor configs, rerankers (e.g., cross-encoders).
- Model Ecosystem: Experience with OpenAI, Azure OpenAI, Anthropic, Cohere, local models (Llama, Mistral) and serving frameworks (vLLM/Triton).
- Guardrails & Safety: Tools like Guardrails.ai, LlamaGuard, content moderation APIs; jailbreak detection.
- Evaluation & Tracing: LLM traces (LangSmith, Phoenix), synthetic eval sets, human-in-the-loop feedback.
- Infra as Code: Terraform/Helm; secrets management (Vault/KMS).
- Streaming & Queues: Kafka/RabbitMQ; event-driven RAG updates.
- Search Platforms: Elastic/OpenSearch integrations; hybrid retrieval pipelines.
- Caching & Cost Controls: Redis, response caching, token usage optimization.
ROLES AND RESPONSIBILITIES:
LLM Ops & Orchestration:
- Operationalize LLM workloads: prompt/version management, model routing, A/B testing, guardrails, and safety checks.
- Design and operationalize agentbased workflows using frameworks like LangGraph, LangChain, Google ADK, or equivalent, selecting appropriate solutions based on usecase complexity, scalability, and enterprise constraints.
- Implement evaluation frameworks (hallucination checks, factuality, toxicity, relevance) and maintain quality SLAs.
- Set up observability: tracing (prompt/response), latency, cost tracking, and drift monitoring.
RAG Pipelines (Design to Production):
- Build end-to-end RAG: ingestion chunking embeddings retrieval reranking generation.
- Optimize context windows, prompt composition, and grounding strategies; use hybrid search (BM25 + dense retrieval).
- Support multi-doc, multi-tenant retrieval with metadata filtering and security-aware access controls.
Vector Database Engineering:
- Design and operate vector stores (e.g., Pinecone, FAISS, Milvus, Weaviate) with proper indexing, sharding, HNSW/IVF configurations.
- Configure ingestion pipelines, upserts, TTL/lifecycle, and backup/restore; tune recall/latency trade-offs.
Python & Platform Engineering:
- Build microservices, APIs, and workers in Python (FastAPI/Flask) with robust error handling and retries.
- Containerize and deploy on Docker/Kubernetes; manage CI/CD, secrets, configs, and environment promotion.
- Integrate with cloud services (AWS/Azure/GCP) for storage, queues, monitoring, and model endpoints.
Security, Compliance & Governance:
- Implement data privacy, PII handling, RBAC, audit logging; enforce enterprise guardrails.
- Apply content filters, jailbreaking protection, and safe output policies.
Collaboration & Documentation:
Partner with Data/ML, Product, and Security teams; produce clear runbooks, architecture diagrams, and API contracts.
Job Classification
Industry: IT Services & ConsultingFunctional Area / Department: Engineering - Software & QARole Category: Software DevelopmentRole: Data EngineerEmployement Type: Full timeContact Details:
Company: Tata ConsultancyLocation(s): Hyderabad
Keyskills: Vector Db LLM Large Language Model Retrieval Augmented Generation Python
Tata Consultancy
\n\nConsolidated Analytics serves clients in mortgage lending, servicing, and capital markets. Its solutions span across the mortgage value chain, which includes: business process services, due diligence, consulting and advisory, real estate asset management, and property valuation services and anal...
Job Listings
We are in open beta version of website, please let us know if any issues, at: info(at)indigojobs.in