TECHNOVA AI — Gaming & Workstation Hardware Intelligence

Designed & developed by ALI RAZA

A Streamlit dashboard and Python MCP server for gaming PC planning, workstation budget planning, a 346-record gaming-laptop dataset explorer and recommendation engine, hardware comparisons, basic compatibility checks, PSU estimation, and MCP tool exposure.

Deploy the dashboard free with Streamlit Community Cloud

1. Open https://share.streamlit.io/ and sign in with GitHub.

2. Select repository ALI-RAZA-AI/TECHNOVA-AI, branch main, and main file app.py.

3. Select Deploy. Streamlit installs dependencies from requirements.txt.

The dashboard is hosted separately from the MCP STDIO server. Streamlit Community Cloud does not expose the MCP server as a web endpoint.

Related MCP server: singularity

Run locally (Windows)

py -m venv .venv
.venv\Scripts\activate
pip install -r requirements.txt
streamlit run app.py

Run the MCP server

python -m technova.mcp_server

The MCP server uses STDIO transport and is intended for an MCP-compatible host/client.

Run tests

pytest -q

Features

• Overview dashboard with component budget visualization.\n- Gaming Laptop Intelligence: explore and filter 346 cleaned laptop records, analyse prices/specification scores/GPUs, compare models and generate workload-aware recommendations.

• Gaming PC Builder for budget planning.

• Workstation Planner for AI/ML, 3D rendering, video editing and general workstation workloads.

• Hardware Explorer with search and category filtering.

• Compatibility Lab for CPU socket, RAM generation and GPU clearance.

• PSU wattage calculator.

• MCP Tools & Demo page and six MCP server tools.

• Dark / Light interface theme switch.

MCP tools

• pc_budget_planner

• psu_calculator

• pc_compatibility_checker

• compare_hardware_specs

• recommend_workload_build

• search_sample_catalog\n- gaming_laptop_recommender\n- gaming_laptop_dataset_summary

Data and limitations

The desktop hardware catalogue is illustrative sample data. The gaming-laptop module uses the uploaded final_cleaned_dataset.csv (346 records, 16 columns), bundled as data/gaming_laptops.csv.gz.b64 to keep the repository compact. The source prices are Indian rupees (INR ₹), not PKR. Recommendations use transparent weighted heuristics; they are not a trained ML model. The dataset is not a live retailer feed. Product profiles and PKR prices are placeholders, not current retail listings. The app does not claim live pricing, stock, benchmark FPS or retailer integrations. Build allocations are heuristics, not verified parts lists. Validate exact models, BIOS support, RAM QVL, PSU connectors, cooling, dimensions and vendor documentation before purchase.

No paid API key is required. An LLM-backed autonomous agent, live retailer adapters and live market feeds are not included in this demonstrator.