AI Architect

LeadSquared; Bengaluru; On-site; INR 3,000,000 - 4,600,000; Full time

Job summary

LeadSquared seeks an AI Architect to design and scale a multi-agent platform powering AI-driven CRM products. You will shape architecture for production-grade systems, balancing reliability, latency, and security, while delivering at scale in a customer-facing role.

You will work on RAG, vector databases, and orchestration across Lumen, Agent Studio, and Invorto-style products, driving practical solutions with hands-on engineering leadership.

Qualifications

• 5–8 years of experience in software/AI engineering with multi-agent platform design and production-grade implementation.
• Strong hands-on Python expertise and cloud/production experience.
• Deep understanding of LLM prompting, evaluation, and deployment considerations.
• Experience with retrieval-augmented generation (RAG) and vector databases.
• Proven track record delivering scalable, reliable AI systems in production.

Responsibilities

• Architect and design multi-agent platforms that operate reliably at production scale.
• Define the technical approach for RAG, vector search, and knowledge grounding across AI products.
• Set standards for LLM orchestration, agent-to-agent communication, memory, and reasoning pipelines.
• Evaluate vector databases and retrieval strategies for different product needs.
• Collaborate with engineering teams to translate architecture into production-grade implementations.
• Drive decisions on reliability, latency, cost, and security for AI workflows.
• Engage with customers and stakeholders to understand requirements and ensure architecture supports real-world use cases.
• Stay hands-on with prototyping, code/design reviews, and solving hard technical problems.

Skills

Python

LLM concepts

System design

Production deployment

Tools

Pinecone

Weaviate

Milvus

FAISS

Job description

We're building the next generation of AI-powered business software. With Lumen, we're reimagining how users interact with CRM through an intelligent interface where they can simply ask AI to take actions, retrieve knowledge, and get work done. With Agent Studio, we're enabling businesses to build and deploy their own AI agents for real-world workflows. And with Invorto, we're bringing AI to voice, powering intelligent voice agents for customer and operational use cases.

This is a chance to work on hard problems across LLMs, agents, orchestration, and reliability, building zero-to-one, owning your area end-to-end, and shipping to production at scale.

Note: This is a customer-facing role, and strong communication skills are essential.

About the Role

We're looking for an AI Architect to help design and scale the multi-agent platform that powers Lumen, Agent Studio, and Invorto. This role sits at the intersection of deep technical architecture and real-world production systems, you'll be responsible for the design decisions that determine how reliably, efficiently, and securely our agentic AI systems operate at scale.

What You'll Do

• Architect and design multi-agent platforms and systems that operate reliably at production scale
• Define the technical approach for retrieval-augmented generation (RAG), vector search, and knowledge grounding across our AI products
• Set architectural standards for LLM orchestration, agent-to-agent communication, tool use, memory, and reasoning pipelines
• Evaluate and select the right vector databases, retrieval strategies, and model choices for different product needs
• Partner closely with engineering teams to translate architecture into production-grade, scalable implementations
• Drive decisions on system reliability, latency, cost, and security for AI-driven workflows used by enterprise customers
• Work directly with customers and internal stakeholders to understand requirements and ensure the architecture supports real-world business use cases
• Stay hands-on - prototyping, reviewing code and designs, and solving hard technical problems alongside the team
• Keep the architecture current with the fast-evolving LLM and agentic AI landscape

What We're Looking For

• 5-8 years of experience in software/AI engineering, with demonstrated experience designing and architecting multi-agent platforms at production grade
• Strong hands-on expertise in Python
• Deep understanding of LLMs - prompting, fine-tuning, evaluation, and production deployment considerations
• Practical experience with RAG architectures and vector databases (e.g., Pinecone, Weaviate, Milvus, FAISS, or similar)
• Proven experience taking AI systems from design to production at scale, with a strong focus on reliability and performance
• Strong system design skills, with the ability to make sound architectural trade-offs in a fast-moving environment