Senior AI Engineer Required – AI Search, RAG & API Development for SaaS Platform

$250-750 CAD

Posted 1 minute ago

•

Ends in 6 days

$250-750 CAD

Paid on delivery

Senior AI Engineer Required – AI Search, RAG & API Development for SaaS Platform We are looking for an experienced AI engineer or specialist development team to build the AI integration and search intelligence layer for a commercial, multi-tenant website search SaaS platform. The platform will enable organizations to add AI-powered search and conversational chatbots to their websites. It will support multiple content sources, including Drupal, website crawlers, sitemaps and APIs. Scope of Work Integrate multiple AI providers, including OpenAI, Anthropic (Claude) and Microsoft Azure OpenAI. Develop a provider-independent AI abstraction layer. Implement secure Bring Your Own Key (BYOK) functionality. Build content processing, metadata extraction and AI classification. Generate embeddings and implement vector and hybrid search using OpenSearch or Elasticsearch. Develop automatic search filter generation based on website content, tags and categories. Implement RAG-based conversational search with accurate source citations. Develop documented REST APIs for chatbot, search, content processing and metadata management. Support English and French throughout the AI functionality. Support public and authenticated content with permission-aware retrieval. Implement AI usage tracking, daily/monthly token allowances and subscription-based limits. Automatically switch to basic keyword-search chatbot mode when AI allowances expire. Integrate usage-exhaustion events with email and SMS notification services. Technical Experience Required Production-grade RAG and AI search implementations. OpenAI, Anthropic and Azure OpenAI APIs. Embeddings, vector databases and semantic search. OpenSearch / Elasticsearch. Python and/or Laravel API integration. Multi-tenant SaaS architecture. Secure API development and access control. Multilingual AI applications. Expected Deliverables Complete AI integration architecture. Working AI provider integrations. Content processing and metadata intelligence engine. Search and chatbot APIs. RAG and hybrid search implementation. BYOK functionality. Usage metering and quota enforcement. API documentation and automated tests. Deployment documentation and complete source-code handover. When Applying, Please Provide Examples of production AI search or RAG projects you have developed. Your proposed technical architecture. Relevant experience with the AI providers listed above. Proposed team structure, if applying as an agency. Estimated timeline and milestone-based cost breakdown. We are looking for someone capable of building a production-ready AI intelligence platform, not simply integrating a basic chatbot. This is part of a larger commercial SaaS project, with potential for ongoing development and long-term collaboration. Please apply only if you have demonstrable experience delivering production AI and search systems.

Project ID: 40806619

About the project

14 proposals

Open for bidding

Remote project

Active 5 secs ago

Place your bid

Bid amount Email address

Benefits of bidding on Freelancer

Set your budget and timeframe

It's free to sign up and bid on jobs

14 freelancers are bidding on average $355 CAD for this job

User Avatar
Flag of INDIA

Hi, I can build your production-grade AI search layer with a provider-agnostic abstraction for OpenAI, Anthropic, and Azure OpenAI, plus secure BYOK, multilingual RAG, vector/hybrid search, and permission-aware retrieval. I’ll design the content ingestion, metadata extraction, embeddings pipeline, REST APIs, quota enforcement, and fallback keyword mode, with citations, testing, and deployment docs. Start the chat so we can discuss the architecture in detail. Regards, Sai Bhaskar

User Avatar
Flag of INDONESIA

Hi there, I'm excited about your project to build a robust AI intelligence and search layer for your SaaS platform. I have extensive experience architecting and deploying production-grade AI search and RAG solutions, including multi-tenant platforms with strong API and access control requirements. Understanding your needs, I propose a modular, provider-agnostic AI abstraction layer supporting OpenAI, Anthropic (Claude), and Azure OpenAI. My approach includes secure BYOK implementation, content ingestion pipelines (crawlers, sitemaps, APIs), and intelligent metadata extraction. I have delivered multilingual, permission-aware search systems with semantic vector search using Elasticsearch and OpenSearch, including hybrid retrieval and RAG-based conversational chat with accurate source attribution. I’ll ensure robust API design for search, chat, and content management, with full documentation and automated tests. Usage tracking, token allowances, quota enforcement, and graceful fallback to keyword-based search will be integral parts of the platform. I’ve previously integrated AI usage metering and real-time notification systems (email/SMS) at scale for SaaS platforms. For similar projects, I have developed enterprise RAG search and chatbot systems for e-commerce and knowledge management SaaS products, leveraging OpenAI/Azure APIs, embeddings, and secure multi-tenant architectures. I am confident in delivering a secure, scalable, and production-ready AI architecture ready for commercial deployment. I look forward to sharing my proposed technical architecture and past project examples. Best Regards

User Avatar
Flag of SERBIA

Hi, The budget and scope don't align: production RAG with multi-provider switching, vector search, BYOK, usage metering, multilingual support and permission-aware retrieval across Drupal, crawlers and APIs is a 3–6 month build for a team, not a $250–750 contract. Either the budget needs to scale to match the full platform, or the scope needs to narrow to a single piece you can ship first. If you want to move forward, I'd propose starting with one concrete phase: build the OpenAI integration and embeddings pipeline for a single content source (say, sitemap crawl), set up OpenSearch indexing, and deliver a working search API with basic RAG. That establishes the architecture and proves the approach before adding Anthropic, Azure, BYOK, multitenancy and the rest. This phase would take 4–6 weeks and cost accordingly. Once it works, each additional provider, content source and feature becomes incremental. Send the current tech stack, any existing crawlers or content pipelines, and which phase makes sense to start with. I'll outline the first milestone in detail. Denis

User Avatar
Flag of BANGLADESH

Hi There! I specialize in AI-powered SaaS development, with 11+ years of experience, 465+ completed projects, and Top 1% verified freelancer status. I can help build a scalable AI search and RAG intelligence layer for your platform. Here’s how I can help: 1. Integrate OpenAI, Claude, and Azure OpenAI through one abstraction layer. 2. Develop secure BYOK and multi-tenant architecture. 3. Implement embeddings, vector search, and hybrid retrieval. 4. Build multilingual RAG chatbots with accurate source citations. 5. Develop REST APIs, usage tracking, quota controls, and automated tests. I'll prioritize security, retrieval accuracy, scalability, and maintainable documentation. Could you share the attached technical specification so I can assess the architecture and milestones?

User Avatar
Flag of NIGERIA

Hello, This project for an AI search and RAG system for a SaaS platform sounds really interesting. I have experience building AI integrations and working with APIs, and I've looked at the requirements for integrating multiple AI providers like OpenAI and Anthropic. My first step would be to map out the provider-independent abstraction layer you've described. I work out of Lagos, Nigeria, and have a background in Python and Laravel, which I believe would be very helpful here. Let me know if you'd like to discuss this further.

User Avatar
Flag of NIGERIA

You need a provider‑independent AI layer that handles multi‑tenant search, RAG chat and secure BYOK for a SaaS platform integrating OpenAI Anthropic and Azure OpenAI. I will build an abstraction API that normalises calls to each provider, implement BYOK key storage with encryption, and create a content pipeline that crawls Drupal sites, parses XML sitemaps and ingests API feeds. I will extract metadata, run classification models, generate embeddings and store them in OpenSearch for vector and hybrid search. I will develop automatic filter generation from tags and categories and a RAG conversational engine that returns precise source citations in English and French. I will deliver fully documented REST endpoints for chatbot interaction, search queries, content processing and metadata management, all containerised with Docker and secured behind Nginx. Could you confirm the preferred authentication method for the external APIs? Let’s chat, lock the exact scope and get started.

User Avatar
Flag of SPAIN

Hi there, we noticed the need for an AI integration and search intelligence layer for your multi-tenant SaaS platform. Our team can develop a provider-independent AI abstraction layer and implement RAG-based conversational search that accurately cites sources, utilizing technologies like OpenSearch to generate embeddings and support search filtering based on your content. We're a small team of developers from Germany and the Netherlands, based in Spain, and we take pride in delivering quality work our clients can rely on with complete peace of mind. Our expertise includes API integrations with AI providers like OpenAI and Azure, ensuring a secure BYOK setup along with comprehensive usage tracking functionalities to meet your requirements. We're confident in delivering a production-grade solution and look forward to collaborating on this project. Regards, Ali Parandeh IT Director

内容来自 Freelancer News,版权归原作者所有。

阅读原文 ↗

把喜欢的内容,留在你的新标签页。

打开我的订阅