AI Orchestration Engineer Needed: High-End Multi-Model "Consensus" Tool ($750 Fixed)
Budget: $250 – $750 CAD
I need an AI expert to build a premium, floating AI Virtual Assistant for three platforms: Proxima Study, ProximaRide, and Canadian Exports. This is NOT a simple API wrapper. It is a "Supreme Judge" system that queries multiple LLMs and synthesizes the best answer.
Technical Workflow:
1- Pre-Filter (The Guardrails): Before any API is called, the system must check the query against site-specific rules (e.g., Proxima Study will NOT do homework or write essays; it will redirect the user to learning resources).
2- The Trio Call: Simultaneously query GPT-4, Gemini, and Claude.
3- The Logic Layer: * Skip Laggards: If a model takes too long, the system moves on without it.
- Discard Anomalies: If one answer is a wild outlier compared to the others, it is ignored.
4- The Synthesis: The remaining answers are sent to a final "Lead Model" (Gemini or Claude) to create one polished, expert response.
5- Usage Caps: Implement a strict cap of 3 queries per hour per user, with an expandable framework to allow for "Premium" paid upgrades.
The Design (Premium Requirement):
>> UI/UX: Must look and feel like a high-end, professional tool. No "default" bot looks.
>> Expandable Window: The chat interface must be able to expand up to 90% of the screen width for deep reading.
>> Wait Animation: While the 4-step backend process happens (up to 30s), the UI must show professional status updates (e.g., "Consulting multiple sources," "Cross-referencing data," "Filtering anomalies").
>> Specs: Inter Font, 12px corner radius, Primary Color #0066FF.
Rules of Engagement:
(a) Budget: $750 Fixed.
(b) Timeline: 7-10 Days.
(c) No Fluff: I have a web developer for the site; I need you for the AI Logic and Premium UI.
(d) Ownership: I am looking for a builder, not a consultant. If you try to overcomplicate the "difficulty" of API calls to justify a price hike, do not bid.
For an extra-pay: Bonus Capability (Voice Interaction): I am interested in adding a 'Talk to the VA' feature where users can speak their questions and hear the answers. I am looking for a developer who can implement a high-end, low-latency voice solution (Speech-to-Text and Text-to-Speech). The voice must sound professional and human, not robotic. If you have experience building voice-enabled AI agents, please include this in your bid as a separate option. I want to know which tools you recommend for the fastest, most premium experience.
Technical Workflow:
1- Pre-Filter (The Guardrails): Before any API is called, the system must check the query against site-specific rules (e.g., Proxima Study will NOT do homework or write essays; it will redirect the user to learning resources).
2- The Trio Call: Simultaneously query GPT-4, Gemini, and Claude.
3- The Logic Layer: * Skip Laggards: If a model takes too long, the system moves on without it.
- Discard Anomalies: If one answer is a wild outlier compared to the others, it is ignored.
4- The Synthesis: The remaining answers are sent to a final "Lead Model" (Gemini or Claude) to create one polished, expert response.
5- Usage Caps: Implement a strict cap of 3 queries per hour per user, with an expandable framework to allow for "Premium" paid upgrades.
The Design (Premium Requirement):
>> UI/UX: Must look and feel like a high-end, professional tool. No "default" bot looks.
>> Expandable Window: The chat interface must be able to expand up to 90% of the screen width for deep reading.
>> Wait Animation: While the 4-step backend process happens (up to 30s), the UI must show professional status updates (e.g., "Consulting multiple sources," "Cross-referencing data," "Filtering anomalies").
>> Specs: Inter Font, 12px corner radius, Primary Color #0066FF.
Rules of Engagement:
(a) Budget: $750 Fixed.
(b) Timeline: 7-10 Days.
(c) No Fluff: I have a web developer for the site; I need you for the AI Logic and Premium UI.
(d) Ownership: I am looking for a builder, not a consultant. If you try to overcomplicate the "difficulty" of API calls to justify a price hike, do not bid.
For an extra-pay: Bonus Capability (Voice Interaction): I am interested in adding a 'Talk to the VA' feature where users can speak their questions and hear the answers. I am looking for a developer who can implement a high-end, low-latency voice solution (Speech-to-Text and Text-to-Speech). The voice must sound professional and human, not robotic. If you have experience building voice-enabled AI agents, please include this in your bid as a separate option. I want to know which tools you recommend for the fastest, most premium experience.