Full-Stack SaaS Developer Needed to Build AI Intelligence Center MVP

We are hiring a strong full-stack developer to build the first working MVP of the Pathwaize Intelligence Center. Pathwaize is an AI-powered operating system for real estate investors. It helps investors answer calls, automate follow-up, manage leads, and simplify operations so they close more deals with fewer disconnected tools. The Intelligence Center is a new standalone dashboard that will live inside GoHighLevel as either a marketplace app or iframe custom menu page. It should feel like an AI-powered execution console, not a generic chatbot or traditional analytics dashboard. The goal of V1 is to build a lean MVP that proves this core workflow: Run engine Generate output Review output Request edits Approve output Push/save approved output. This is not a speculative design project. We already have a detailed product brief, feature scope, UI direction, and drafted engine workflow files. We need someone who can turn that into a clean, functional, production-minded MVP. V1 SCOPE: The Lean MVP should include: Core Intelligence Center UI (mockup image attached) Intel Brief UI stub Knowledge Engine Lite Authority Engine wired end-to-end Newsletter Engine wired end-to-end Outputs Pending Review Prompt chips Persistent intelligence input Review, approval, revision, and save/push workflows Basic backend engine orchestration Backend-based LLM routing layer V1 ENGINES: The MVP will include three primary AI engines: Knowledge Engine Lite — generates, reviews, revises, approves, and saves the customer’s core business knowledgebase to approved storage destinations such as Google Drive or GitHub. Authority Engine — generates founder-authority content outputs, saves them for review, supports revisions, and allows approved content to be pushed or saved to the intended destination. Newsletter Engine — generates newsletter drafts, subject line options, preview text, review-ready outputs, revision workflows, and approval-gated save/push actions. We already have drafted workflow / skill files for these engines. These files should be treated as starting points, not final production code. The selected developer will need to review them, adapt them to this project, connect them to the backend engine orchestrator, and modify them to fit the Intelligence Center’s run, review, revise, approve, save/push workflow. Intel Brief Stub: The Intel Brief should be included in the UI as a lite/manual/stubbed section. It should show the intended structure and placeholder/example items such as: What changed What is working What is leaking What needs action Recommended decisions Priority list Full analytics automation is not required for V1. REQUIRED TECHNICAL FOUNDATION: The MVP should include: Pathwaize backend / engine orchestrator Backend-based LLM routing layer Prompt/template management Engine run records Output storage Output status tracking Review and approval workflow Basic revision workflow Basic usage and cost logging Basic error handling Google Drive connection for KB storage GitHub connection for KB storage if selected as KB destination Subaccount/workspace association where needed LLM ROUTING: The frontend should not call OpenAI, Claude, or any LLM provider directly. The frontend should communicate with the Pathwaize backend. The backend should handle prompt execution, model/provider selection, engine orchestration, permissions, logging, usage tracking, cost tracking, run history, error handling, and approval-gated actions. The system can start with one primary LLM provider in V1, but it should be structured so additional providers or models can be added later without rebuilding the product. Examples of backend-routed actions may include: Running the Knowledge Engine Running the Authority Engine Running the Newsletter Engine Revising an output Summarizing an engine run Classifying a user command Saving outputs for review Preparing approved outputs to be pushed or saved LLM-generated outputs should not automatically publish, send, or update critical systems without user approval. IDEAL CANDIDATE: You should be comfortable building practical SaaS MVPs with clean UI, backend workflows, API integrations, LLM workflows, and user approval flows. You should have experience with some or all of the following: Full-stack SaaS development React / Next.js or similar frontend frameworks Backend API development LLM integrations through a backend routing layer OpenAI / Claude / model routing concepts Google Drive API GitHub API GoHighLevel apps, custom menu links, marketplace apps, or iframe-based integrations Database design for run history, output storage, statuses, and user/workspace records Clean UI implementation from a written product brief Turning existing prompts, workflows, or AI skill files into functional SaaS product features WHAT WE DON'T WANT: Please do not apply if you only build simple chatbot wrappers. This product needs to feel like an AI workflow execution console where users can run engines, review outputs, approve work, request revisions, and save or push approved assets. We do not want the V1 overbuilt with advanced analytics, future engines, continuous KB monitoring, proactive gap detection, or unnecessary complexity. V1 should be focused, useful, and built to validate the core product workflow quickly. PRIORITY BUILD ORDER Build the UI shell Wire the Authority Engine end-to-end Wire the Newsletter Engine end-to-end Build Knowledge Engine Lite for KB generation and storage Add Outputs Pending Review Add approve / request edits / re-run flow Add Intel Brief UI stub Add future-engine placeholders only if helpful COMMUNICATION NOTE: Please keep all communication inside Upwork during the application and interview process. Do not contact us directly by email, phone, LinkedIn, Facebook, Instagram, or any other channel unless we specifically request it. Candidates who bypass this process may be removed from consideration. We are looking for someone who can follow instructions, communicate clearly, and work through an organized hiring process. TO APPLY: Please include: Examples of SaaS dashboards, AI apps, or workflow tools you have built Your recommended tech stack for this MVP (currently using AWS and Supabase) Your experience with LLM-backed workflows Your experience with GoHighLevel integrations How you would structure the backend LLM routing layer so the product is not hard-coded to one model or provider How you would prevent approval-gated actions from publishing, pushing, or updating connected systems without user approval How you would approach building this as a lean MVP without overengineering it Your estimated timeline and fixed-price or milestone-based budget range I am looking for someone who can think clearly, communicate directly, and build a functional MVP that is clean, scalable enough for V1, and not bloated. Strong candidates will understand that the goal is not to build every future idea. The goal is to ship the first usable version that proves the Intelligence Center workflow and creates immediate value for operators.