Selected Projects

ATS Forge is a production-grade AI-powered CV builder that generates ATS-optimised resumes tailored to specific job descriptions. Users can paste a JD or JD URL and either upload an existing CV or complete their profile using the AI-assisted “Enhance with AI” buttons. ATS Forge AI then analyses the role, plans a targeted CV strategy, and generates a professionally formatted, keyword-optimised resume with intelligent skills categorisation, rewritten experience bullets, and export options in PDF and DOCX formats.

The system is built on a dual-LLM architecture using Groq (Llama) as primary for speed and OpenAI GPT-4o as fallback for quality, with a multi-stage pipeline covering JD analysis, strategy planning, content generation, and document formatting — all deployed as a production Flask application.

Sheet Insight AI is a production AI-powered spreadsheet analyst that connects to Google Drive and lets users ask plain-English questions about their data. It converts natural language queries into SQL, executes them against the actual spreadsheet data using DuckDB, and returns exact answers — totals, trends, rankings, date-range comparisons — with structured tables and clear explanations.

The system supports Google Sheets, CSV, and Excel files with multi-sheet tab switching. Built on a dual-LLM architecture using Groq as primary and OpenAI as fallback, with a modular prompt system for SQL generation, post-processing, and result interpretation — all with zero data storage and read-only Google Drive access.

Ad Craft AI is a production-grade AI ad creative generator that produces 3 professional ad variations from a single brief. Users paste any product URL for automatic extraction of brand, product, description, and category — or enter details manually. Upload a logo and product images, choose from 8 ad styles and 20+ format presets covering every major platform (Instagram, Facebook, YouTube, TikTok, Pinterest, LinkedIn, Twitter/X, Google Display), and get campaign-ready creatives with a single click.

Two generation modes are available: Product Composite generates an AI background scene and composites product images and logo pixel-perfect via PIL, while AI Product Studio uses GPT Image to recreate products directly inside the scene. The system features GPT-4o-mini vision analysis for colour-aware compositions, a three-tier content safety system, campaign intelligence with funnel-aware CTAs, and per-concept layout controls across all 3 variations.

Designed and built multiple LLM-powered assistants for conversational search, learning support, and problem-solving, combining retrieval-augmented generation with multi-step reasoning and agentic control flows. Used an MCP-based orchestration layer to manage tools, context, and memory across assistants.

These systems integrated structured knowledge sources with LLMs to provide grounded responses while supporting follow-up questioning, exploration, and clarification rather than one-shot answers. Several assistants were deployed in real usage contexts, while others served as experimental platforms for interaction research.

Design and implementation of a full-stack mental-health prediction system integrating a machine-learning model with a user-centred interface for patients, clinicians, and administrators. Operating in a safety-critical domain, the platform provided predictive insights while explicitly avoiding full automation.

The system architecture prioritised transparency, staged feedback, and role-aware information access, ensuring AI outputs supported—rather than replaced—clinical judgement.

ConnectCircle is a next-generation social media platform built around a global chat system with enterprise-grade security and 24/7 admin control. Unlike traditional platforms, every interaction can be monitored and reported through global safety windows that enable continuous moderation, abuse prevention, and real-time intervention.

Privacy is not optional—it's the default. Users can participate in global chat, but private messaging is restricted to friends, and only mutually revealed friends can view each other's posts. Identity and content access are unlocked strictly through explicit, mutual consent.

DeepGrade AI is an AI-assisted grading and feedback platform designed to support students, teachers, and administrators in educational assessment workflows. The system applies large language models to evaluate handwritten and structured student responses against defined rubrics, generating scores and explanations while keeping humans firmly in control of final decisions.

Operating in a multi-role environment, AI outputs are advisory rather than authoritative, ensuring alignment with real-world educational practices. A core focus was understanding how users interpret AI-generated feedback through interaction-log analysis and qualitative evaluation.