Associate Architect - Machine Learning

Quantiphi; Bengaluru; On-site; INR 2,100,000 - 3,600,000; Full time

Job summary

Quantiphi seeks a hands-on Generative AI Engineer to join the AI Capabilities team in Bengaluru. You will design and implement end-to-end GenAI pipelines, including data ingestion, embedding selection, vector indexing, retrieval logic, and deployment as API endpoints.

You will build agentic architectures with routing, memory management, and cross-LLM orchestration, while enforcing safety, governance, and observability.

Qualifications

• Hands-on experience designing and deploying large-scale GenAI pipelines including data ingestion, embedding selection, vector indexing and retrieval logic.
• Experience building multi-agent architectures with routing, memory management, and summary generation across LLMs.
• Practical knowledge of GenAI ops: PEFT/LLM fine-tuning, evaluation pipelines, and RAG configuration management.
• Familiarity with AI safety, governance, and compliance requirements in regulated environments.
• CI/CD and infrastructure as code practices (GitHub Actions, Jenkins, Terraform) for GCP-based AI infra.

Responsibilities

• Develop and deploy GenAI capabilities as pipelines, agentic workflows, and REST APIs.
• Design and implement agent-to-agent communication, routing models, and memory/state management with observability.
• Build, monitor, and improve evaluation, safety guardrails, and AI governance across services.

Tools

BigQuery Vector Search

GKE

Cloud Run

Pub/Sub

BigQuery

LangChain

LlamaIndex

Vector embedding design

Model routing & rate limiting

Observability tooling

Job description

Role Summary

We are seeking a hands-on and technically strong Generative AI Engineer to join our AI Capabilities team.

Required Skills

• pipelines, including data ingestion, chunking strategy design, embedding selection, vector indexing (e.g., BigQuery Vector Search), retrieval logic, and deployment as API endpoints. Strong understanding of RAG evaluation metrics (Faithfulness, Answer Relevancy, Context Precision/Recall) and continuous knowledge base updating pipelines.
• Agentic Architecture & Implementation: Demonstrated experience building multi-agent systems, including Semantic Router, Agent Orchestrator (with workflow management), Stateful Orchestration Runtime (Reasoning Engine), and Session State/Memory (LTM/STM) management. Ability to implement agent-to-agent communication protocols, intent recognition, routing models, and agent handoff mechanisms with summary generation. solutions covering model routing, rate limiting, caching, observability, fallback logic, and policy enforcement across multiple LLM providers. Familiarity with Tool & Integration Runtime (API calls, MCP, A2A) and Event Bus/Messaging architectures for asynchronous, decoupled AI service coordination.
• GenAIOps & MLOps Frameworks: Hands-on experience implementing GenAIOps practices including Prompt Engineering, RAG configuration management, embedding lifecycle management, PEFT/LLM fine-tuning, Prompt Registry versioning, and LLM evaluation pipelines. Solid understanding of MLOps principles covering model training, validation, experiment tracking, model registry, serving, monitoring, and explainability.
• AgentOps Implementation: Experience building and operationalizing AgentOps frameworks for developing, deploying, monitoring, and governing AI agents, including scenario testing, approval gate workflows, memory management, tool call tracking, and latency/success rate monitoring.
• GCP AI/ML Platform Proficiency: Strong, hands-on expertise with GCP services critical to AI Cloud Run, GKE, Cloud Storage, Pub/Sub, and BigQuery. Ability to deploy GenAI capabilities as pipelines, agentic workflows, REST APIs, and automation scripts. Experience deploying AI services as scalable API endpoints with appropriate authentication, rate limiting, and monitoring.
• AI Safety, Governance & Compliance: Practical experience implementing AI safety guardrails, output filtering, PII protection, bias detection, and audit logging within GenAI platforms. Understanding of data sovereignty requirements and compliance standards relevant to a regulated financial services environment.
• CI/CD & Infrastructure as Code: Experience integrating GenAI capabilities into CI/CD pipelines (GitHub Actions, Jenkins, or Google Cloud Build) for automated testing, evaluation, and deployment. Working knowledge of Terraform for provisioning GCP-based AI infrastructure.

Nice-to-Have

• Experience building AI platform capabilities in GCP cloud environment, ideally supporting a "build once, leverage everywhere" reusability model across multiple LBUs.
• Experience with Knowledge Graph architectures integrated with RAG for enterprise semantic discovery and relationship-based reasoning.
• Familiarity with RAG orchestration frameworks such as LangChain or LlamaIndex, and LLM evaluation toolsets such as RAGAS, DeepEval, or Vertex AI Rapid Eval.
• Experience with Context Store, Vector Store, Embedding infrastructure, and Feature Store design as components of an AI-ready data layer.
• Knowledge of the financial services or insurance (BFSI) domain, including data sovereignty, regulatory compliance, and risk management requirements across APAC markets.
• Experience working within large-scale enterprise programs involving multiple implementation partners and formal governance and change