DevNexa.
Digital Product Engineering Studio

We build digital products that move businesses forward.

System Status● WebGL & 3D Shaders Initializing
0%

Data Analytics

Team Experience

MyTaskMate AI

A database-backed productivity assistant with semantic search, built during an AI/ML internship at C-DAC Pune.

Client

C-DAC Pune

Industry

Productivity / AI Tools

Services

AI/ML Development

Status

Completed

The Challenge

Keyword-only search made it hard to retrieve relevant tasks and notes quickly; the assistant needed structured storage plus retrieval that understood meaning, not just exact matches.

The Solution

Ajay integrated PostgreSQL for structured storage, implemented semantic retrieval so information could be found by meaning, and optimized search with FAISS for significantly faster lookups.

Team Contribution

  • PostgreSQL-backed structured data storage
  • Semantic retrieval implementation
  • Semantic search optimization with FAISS
  • AI assistant and LLM-based workflow design (LangChain)

Key Features

  • Semantic (meaning-based) information retrieval
  • FAISS-optimized search, ~3x faster retrieval
  • PostgreSQL-backed task and information storage
  • LLM-assisted productivity workflows

Technology

PythonPostgreSQLFAISSLangChainLLMs

Results

MyTaskMate AI achieved roughly 3x faster information retrieval through FAISS-optimized semantic search.

Have a similar project?