Build Router Layer for AI Islamic App (RAG + Multi-Module System)

@blackrockwise

Hi there, I like how you have outlined your project description with clear milestones and expectations. You need a developer to build a Router Layer for your AI Islamic App that accurately classifies user queries into specific modules and integrates smoothly with your existing FastAPI backend and /ask endpoint. The task includes query classification, normalization to improve retrieval, routing to the appropriate modules, and ensuring consistent behavior without hallucinations. With extensive experience in Python backend development and API integration, I specialize in building clean, scalable control systems using rule-based and AI classification methods. I understand the importance of query normalization for improving accuracy and can deliver a reliable router layer within your 1–2 days timeline. My approach will combine a lightweight, rule-based classifier tailored to the Islamic domain glossary you provided, ensuring precise routing and seamless integration with placeholders for future modules. I will also provide basic testing with sample queries to guarantee functionality. I am excited about the opportunity to contribute to HalalyAI and support your ongoing development needs. Please feel free to reach out so we can discuss your requirements in detail and get started promptly.

@avinashdubeyg209

Noticed that your existing Islamic RAG system needs a flexible Router Layer to streamline user queries. With recent experience on a similar project using FastAPI, I'm well-equipped to classify and route queries efficiently across your modules. Curious about the intricacies of each module—do any require custom logic, or is this a straightforward routing task? Can start today and can walk you through the initial approach. Let me know if you're ready to dive deeper.

@chiragardeshna

Hi there, Strong alignment with this project comes from experience building modular AI systems with routing layers, query preprocessing, and RAG integrations. Clear understanding of the requirement to implement a FastAPI-based router with classification, normalization, and controlled routing to multiple modules. Hands-on expertise with Python and LLM workflows ensures clean rule-based or hybrid classification, efficient normalization, and stable integration with your existing /ask endpoint. Risk is minimized through simple, scalable logic, fallback handling, and testing across sample queries. Available to start immediately can deliver within 1–2 days and open for long-term work. Recent work: https://www.freelancer.com/u/chiragardeshna Regards Chirag

@arslanshahid386

Hi there, I see that you're looking to build a Router Layer for your Islamic app to effectively classify and route user queries to the right modules. With 4+ years of experience in backend development using Python and FastAPI, I can help you implement a clean and efficient solution. My approach would involve creating a rule-based classification system to categorize incoming queries and apply normalization before routing them to the appropriate endpoint. I understand the importance of maintaining the existing functionality of your RAG system while ensuring no hallucinations occur. I’ve worked on similar projects that required precise query handling and module integration, which means I can deliver a solid implementation tailored to your needs. Could you clarify how you envision the placeholder logic for the zakat module and other modules to function during initial integration? Best regards, Arslan Shahid

@HemantManglani01

Hi there, I'm Hemant Manglani, a Senior Python Backend Engineer. I've spent the last 5+ years building scalable systems with FastAPI, Django, and AWS. I enjoy taking projects from the initial idea all the way to deployment and beyond, ensuring everything is stable and easy to maintain. Your need for a robust router layer for HalalyAI, especially the query classification and normalization, sounds right up my alley. I can implement this efficiently using a **rule-based classification** system in FastAPI, ensuring precise routing to your modules and proper query normalization before hitting your RAG. This approach offers quick, predictable results within your existing setup. Based on the scope, I estimate this can be delivered within **1.5 days**, keeping the code clean and ready for future modules. I'm keen to help you build this critical component and potentially assist with future stages of HalalyAI. Let's quickly connect to discuss how we can get this integrated seamlessly. Best, Hemant

@cindyviorinar

Hi Client, I’ve read your description and understand you need a lightweight router layer that classifies incoming questions and routes them to the correct module before the existing /ask endpoint. I’ll implement a clear, maintainable FastAPI middleware or endpoint that performs classification (rule-based with opportunity to plug in an AI classifier later), applies normalization for the islamic_rag route, and triggers the appropriate module flows as placeholders where required. I have solid experience building NLP routing and preprocessing layers in Python/FastAPI and integrating with RAG systems. My approach will define a concise rule-based classifier tuned to your module keywords, build a normalization mapping for the islamic_rag path, ensure calls to the existing /ask keep current behavior (preserving “I don’t know” fallbacks), and add unit tests with sample queries. I’ll structure the code for easy extension to AI classification in future. I can deliver this in 1-2 days with clean, documented code and basic tests. Which sample queries or edge cases should I prioritize in testing the router (e.g., multi-intent queries or short ambiguous phrases)? Best regards, Cindy Viorina

@indreshpnk

I’ll implement a lightweight router using a hybrid approach—rule-based classification (fast + predictable) with optional LLM fallback for edge cases. I’ll add a query normalization layer (dictionary + stemming) before routing to your existing RAG /ask endpoint, ensuring no change in current behavior. Clean FastAPI integration, modular design, and test cases included. Delivery within 1–2 days. Experience building RAG pipelines and API routing systems.

@abhiramukk

⭐⭐⭐⭐⭐ Transforming Your AI Islamic App: Router Layer Implementation Hi there, Striving to seamlessly route user queries? HalalyAI's next step involves implementing a crucial Router Layer before the existing /ask endpoint. ✅ How I will help: ↪️ Efficiently build a Query Router to classify user questions into distinct modules, ensuring precise routing logic and query normalization for enhanced retrieval. ✅ Work Experience: *️⃣ Proficient in Python (FastAPI backend) with a knack for seamless integrations and clean, scalable code. ⚫Quick Questions: ❔ How do you envision the scalability of the Router Layer as additional modules are introduced? ❔ Have you considered the potential impact of query normalization on user experience and system efficiency? Let's discuss further, Regards, ✍️ AB

@Sadiyasyeda24

I saw your project and am confident I can deliver on this. I'm currently working on a similar project and have experience in building query routers. Implementing the Router Layer for your AI Islamic App is crucial for routing user queries effectively to the correct modules. With my expertise, I assure seamless integration, maintaining current RAG behavior, and ensuring accurate module triggers. This component will significantly enhance the user experience and streamline the functionality of your app. I invite you to view my portfolio, which showcases the quality and results of my past work. I look forward to hearing from you. Regards, Sadiya

@surjeetkaran

Hi, I’ve gone through your requirement carefully, and I really like the structured approach you’re taking with HalalyAI. Since your RAG system is already in place, I’ll focus on building a clean and reliable router layer that fits seamlessly into your existing FastAPI backend. My approach: I’ll implement a hybrid classification system: Start with rule-based intent detection (fast, predictable, controllable) Keep it extensible so we can later plug in ML/NLP if needed For query normalization, I’ll create a lightweight mapping layer (dictionary + preprocessing pipeline) to standardize inputs before hitting the RAG endpoint—ensuring better retrieval and maintaining your strict output control. Routing will be modular and clear: Direct function triggers for zakat, halal scan, etc. (placeholders) Safe pass-through to /ask only for islamic_rag Clean fallback handling for general_chat I’ll keep the code minimal, readable, and scalable, so future modules can be added easily. Timeline: 1–2 days (can deliver within 48 hours) Experience: Worked with FastAPI APIs, routing layers, and NLP preprocessing in similar structured backend systems. If this aligns with your vision, I’d be happy to start right away and also support future modules. Let’s connect and get this implemented smoothly.

@ahmadraza001912

With over 4 years of expansive experience in backend development and API architecture, I am confident that I am the best fit for your project. My forte lies in Python - Django and FastAPI being my top expertise, which align perfectly with your current tech stack. Drawing from my background, I can bring exceptional value to the table by building a robust router layer as per your requirements and ensuring seamless integration with your existing setup. To address the critical point of query normalization, I suggest employing an AI-based approach using Natural Language Processing (NLP) algorithms, which can significantly improve retrieval accuracy. This also opens avenues for better scaling opportunities in the future, should you choose to expand on this system. Additionally, my skills include working with third-party APIs, which will help me integrate modules such as zakat_flow or prayer_time without a hitch. I understand the importance of clean, simple code for a lightweight but crucial component like this router layer. My work ethic revolves around delivering production-safe code with minimal disruption to live systems. Rest assured about application structure and efficiency – I prioritize performance and scalability in all my projects. Top that off with my proactive communication style and documented deliveries, you have a recipe for successful ongoing collaboration.

@Thiran003

Hello, I can implement the Router Layer + Query Normalization for your Islamic RAG system with clean, scalable FastAPI integration. Approach: • Hybrid classification (rule-based + lightweight NLP for accuracy) • Query normalization using mapping dictionary + preprocessing layer • Modular routing to each component (RAG, zakat, halal, etc.) • Safe fallback handling to maintain strict RAG behavior Includes full integration, testing with sample queries, and clean documented code for future modules. Ready to start immediately. Best regards, Thiran360AI Team

@anuragj62

I think I’m a really good fit for this because I’ve already worked on similar AI + routing style problems. I’ve built agent-based systems and automation workflows (using tools like LangGraph and n8n), where routing queries and controlling flow between modules is a core part of the system. I’m pretty comfortable with Python + FastAPI, and I’ve worked on things like guardrails, fallback logic, and structured AI outputs — which is very similar to what your router layer needs to do. For this, I’d probably go with a clean rule-based classifier first (fast and reliable), plus a solid normalization layer so your RAG gives better results. I’ll keep everything simple, modular, and easy to extend later for zakat, halal scan, etc. I can get this done in 3 days without overcomplicating things

@ratnams7

Approach: Employing a combination of rule-based and AI classification methodologies, I will ensure that relevant queries are effectively routed to the correct module in your Islamic App. By utilizing my expertise in clean and scalable code, I can integrate the Router layer with your existing backend without affecting its performance. I understand the weight and role of this module, thus assuring you not just basic testing but a comprehensive end-to-end integration with your current RAG system. Experience: I've had hands-on experience in similar projects that involved building intricate API systems. Specifically, my work on developing automated trading systems and creating intelligent voice assistants has honed my skills aligning well with the challenge at hand. Having worked on such intricate projects signals my capability to deliver under tight timelines while maintaining high standards of quality. Estimated Time: Given the scope and complexity, I estimate completing this assignment within 1-2 days.

@sonaligade07

I have around 1.5 years of experience in data engineering and analytics, with strong skills in Python, SQL, and data processing frameworks such as PySpark and Databricks. In my current role, I have worked on building data pipelines, handling large datasets, and generating insights that support business decision-making. Additionally, I have hands-on experience with data visualization tools like Power BI, where I have developed dashboards to analyze customer behavior and performance metrics. I have recently enhanced my skill set by working on projects involving data analysis, feature engineering, and basic machine learning concepts. I am highly motivated to transition into a Data Scientist role where I can apply analytical thinking and statistical techniques to solve real-world problems.

@andersond73

Hello, As I am a Python backend engineer, I can do your router layer for HalalyAI in 2 days for 3000 INR. The main risk here is not just classifying the query, but routing it without breaking your current Islamic RAG behavior or sending the wrong intents into /ask. I would build this as a small RouterService in FastAPI with clear intent rules for islamic_rag, zakat_module, halal_scan, prayer_times, qibla_compass, and general_chat, then add a normalization map before RAG so terms like angry→anger, interest/usury→riba, and washing before prayer→wudu are standardized before retrieval. For this scope and budget, I would use a rule-based classifier first, with clean extension points so you can later swap in AI classification if needed. I will integrate it before /ask, keep JSON responses consistent, and include sample-query tests for each route. I have worked on FastAPI, AI workflow, and routing-style backend systems before. Thank you.

@srdank001

Hi, there I’ve worked on FastAPI-based AI systems with routing and preprocessing layers, especially for RAG pipelines where correct classification and normalization are critical. For your router, I’d use a hybrid rule-based approach (keyword + intent mapping) for speed, control, and zero hallucination risk. Each incoming query will first pass through a lightweight classifier that maps it to one of your modules. For Islamic queries, I’ll add a normalization layer using a dictionary-based synonym mapper (e.g., “interest” → “riba”, “mad” → “anger”) before sending it to your existing /ask endpoint. The router will act as a clean middleware service in FastAPI, ensuring each module is triggered correctly while keeping your current RAG behavior intact. Placeholder handlers will be added for zakat, halal scan, prayer times, and qibla. The code will be simple, modular, and easy to extend later with ML-based classification if needed. I can complete this within 1–2 days. I’d be glad to review your current /ask flow—how are you currently handling fallback responses?

@surajr674

Hello, I have reviewed your requirement for building the Router Layer for your Islamic AI system, and I can implement this efficiently within your timeline. ### Approach: I will build a clean and scalable query routing system using a **rule-based + extendable structure**. • **Query Classification:** Keyword + intent-based classification to route queries into modules like islamic_rag, zakat_module, prayer_times, etc. • **Query Normalization:** A mapping layer to improve RAG performance (e.g., interest → riba, donation → sadaqah, mad → anger). • **Routing Logic:** * islamic_rag → normalized → existing /ask endpoint * zakat_module, halal_scan, prayer_times, qibla_compass → placeholder handlers * general_chat → fallback response • **Integration:** Seamless integration with your FastAPI backend while preserving current RAG behavior (no hallucination, proper fallback). • **Testing:** Basic testing with multiple queries to ensure correct routing. --- ### Skills: Python, FastAPI, API Integration, NLP Basics --- ### Timeline: I can complete this within **1–2 days**. --- I am reliable and available for future work (Zakat module, Halal scanner, etc.). Looking forward to working with you. Best regards, Suraj Kumar