5820 Senior Ai Architect Llm Rag Ai Agents jobs near me

Experience: 5+ Years Overall Technical Experience
Relevant AI Experience: 2–3+ Years Preferred

We are a growing multi-store e-commerce business looking for a highly skilled Senior AI Architect to help design and build the knowledge, memory, and governance foundation for our AI agents.

Our goal is to build AI agents that can use company knowledge to analyze problems, create plans, provide recommendations, and support business operations.

However, the AI should not simply search documents and generate answers.

Before providing recommendations or taking approved actions, the AI should be able to check relevant company policies, previous decisions, active plans, permissions, and restrictions.

If the AI finds conflicting information, it should be able to:

• Identify the conflict
• Explain the issue clearly
• Cite the relevant source
• Avoid unauthorized actions
• Escalate the issue for human review when required

We are looking for someone who can think beyond a basic chatbot or simple document-based RAG system.

This role requires strong experience in LLMs, AI Agents, RAG, data architecture, knowledge systems, APIs, governance, and production AI systems.

What You Will Do1. Design Our AI Knowledge Architecture

You will help us design how company knowledge should be structured and made available to AI agents.

This may include:

• Company policies
• SOPs and workflows
• Business decisions
• Project information
• Historical decisions
• Drafts and proposals
• Informal notes
• Restricted or confidential information

You will define how these different types of information should be stored, classified, updated, retrieved, and governed.

2. Build a Reliable Knowledge & Retrieval System

You will design or guide the implementation of systems involving:

• Retrieval-Augmented Generation (RAG)
• Vector databases
• Structured databases
• Metadata
• Semantic search
• Hybrid retrieval
• Knowledge graphs where appropriate
• Document ingestion pipelines
• Context management
• Persistent AI memory

The objective is to ensure AI agents retrieve information that is not only relevant, but also current, authorized, and reliable.

3. Design AI Agent Workflows

You will design AI agents and workflows that can:

• Access approved company knowledge
• Use APIs and external tools
• Retrieve relevant information
• Validate information before responding
• Check policies and restrictions
• Identify potential conflicts
• Request human approval when necessary
• Maintain records of important actions and decisions

4. Build Conflict Detection & Safety Checks

One of the most important parts of this role is designing how AI agents should handle conflicting information.

For example:

An AI agent may find that a proposed action conflicts with:

• An existing company policy
• A previous business decision
• An active project plan
• A security or access restriction
• A more recent version of a document

The AI system should be able to identify the conflict and determine which information has higher authority.

The system should support:

• Source citations
• Authority hierarchy
• Version checking
• Conflict detection
• Human review
• Authorized overrides
• Audit logs

5. AI Governance & Access Control

You will help define how company knowledge should be governed.

This includes:

• Source ownership
• Authority levels
• Access permissions
• Confidentiality levels
• Version control
• Review dates
• Knowledge updates
• Human approvals
• Manual overrides
• Audit trails
• Data retention

6. Design a Vendor-Neutral Architecture

We do not want our company knowledge and AI workflows locked into a single AI provider.

The architecture should allow flexibility to work with providers such as:

• OpenAI
• Claude
• Gemini
• Other future LLM providers

Our company knowledge, governance rules, metadata, permissions, and historical decisions should remain independent from any individual AI model wherever practical.

Required Skills & ExperienceAI & LLM Experience

You should have hands-on experience with:

• Large Language Models (LLMs)
• Generative AI applications
• AI agents or agentic workflows
• Retrieval-Augmented Generation (RAG)
• Prompt and context management
• Tool calling and API integration
• AI evaluation and testing

Data & Knowledge Systems

Experience with some or most of the following:

• Vector databases
• Semantic search
• Structured and unstructured data
• Document ingestion
• Metadata design
• Knowledge management
• Hybrid search
• Knowledge graphs
• SQL and/or NoSQL databases

Technical Skills

Strong experience with:

• Python
• APIs
• System integrations
• Databases
• Backend or cloud architecture
• Workflow automation

Experience with technologies such as LangChain, LangGraph, LlamaIndex, OpenAI APIs, Anthropic APIs, Google AI, FastAPI, vector databases, or equivalent technologies is preferred.

Preferred Experience

Candidates with experience in the following areas will be strongly preferred:

• Production AI systems
• Enterprise RAG systems
• AI copilots
• Internal AI assistants
• Multi-agent systems
• AI governance
• AI observability
• AI evaluation frameworks
• Access control and permissions
• Human-in-the-loop workflows
• Data security and privacy

Experience with e-commerce, Shopify, Slack, Google Workspace, helpdesk systems, CRMs, or business workflow automation is a plus.

What We Are NOT Looking For

This position is not suitable for candidates whose experience is primarily limited to:

• Basic Prompt Engineering
• Simple ChatGPT integrations
• Basic chatbots
• PDF question-and-answer applications
• Simple document uploads to a vector database
• No-code AI automation only

We are looking for someone who understands how to build a scalable and reliable AI system, including:

Knowledge → Metadata → Permissions → Retrieval → Validation → AI Agent → Human Approval → Audit Trail

Ideal Candidate

The ideal candidate has a strong background in software engineering, AI/ML, data engineering, or technical architecture and has spent the last few years working hands-on with modern AI technologies such as LLMs, RAG, and AI agents.

You do not need to have 10 years of Generative AI experience.

However, you should have proven experience building real AI systems and strong overall technical architecture experience.

We are especially interested in candidates who can clearly explain:

• How an AI knows which information to trust
• How knowledge should be classified and versioned
• How conflicting information should be handled
• How permissions should be enforced
• When human approval is required
• How AI decisions can be traced and audited
• How to avoid dependence on a single AI provider

If you have built systems like this and enjoy solving complex AI architecture problems, we would like to hear from you.

Pay: ₹60,000.00 - ₹100,000.00 per hour

Application Question(s):

• How many years of hands-on experience do you have building Generative AI, LLM, RAG, or AI agent systems?
• Have you personally designed or implemented a RAG or enterprise knowledge retrieval system for a real-world or production environment?
• Have you built or designed AI agents that use APIs, tools, external data, workflows, or company knowledge to complete tasks?
• What is your level of hands-on Python experience for AI, APIs, backend systems, or data processing?
• Please briefly describe one AI, LLM, RAG, or AI agent system you personally designed or built. What was your role, and how did the system retrieve and manage knowledge?
• How would you structure company policies, SOPs, project decisions, drafts, informal notes, and confidential information so an AI knows what information it can trust and access?
• If an AI finds that a proposed action conflicts with an existing company policy or previous decision, how would you design the system to detect, explain, escalate, and record the conflict?
• Briefly explain how you would design this workflow: Company Knowledge → Metadata → Permissions → Retrieval → Validation → AI Agent → Human Approval → Audit Log.

Work Location: Remote