A growing collection of software we've designed, engineered, and shipped across products, AI, systems, automation, and infrastructure. Project names are working titles where client confidentiality applies.
Project names shown here are working titles used to describe the work.
Bodh is a private, offline semantic search layer for organisational knowledge. It uses autoencoder-based representation learning to understand queries by meaning rather than exact wording, letting teams search internal documentation, tools, and practices without sending any data to an external AI service.
View case study
Drishti is a competitive-intelligence platform built for a US used-car dealership. It continuously monitors competitor inventory, compares similar vehicles, studies dealership-level patterns, and turns changing market conditions into practical pricing and merchandising recommendations.
View case study
Visionary Extractor uses computer vision, image processing, and OCR to turn packaging images into structured product information, supporting cataloguing, inventory, verification, warehouse operations, and supply-chain workflows.
View case study
Shodh is a research assistant built for a difficult operating reality: questions arrive in unpredictable ways, while the available knowledge remains fixed and curated. It absorbs incoming questions, groups them intelligently against indexed research, sends them through retrieval workers in efficient batches, and protects the system when demand exceeds safe capacity.
View case study
Vega is a high-throughput inference serving system designed for extreme traffic bursts on fixed private infrastructure. Instead of relying on reactive autoscaling, it uses shared queuing, dynamic batching, backpressure, and controlled load shedding to maximise existing compute utilisation.
View case study
Atithi is an AI-powered platform that helps hotel teams understand guest feedback beyond simple star ratings or positive/negative labels by performing aspect-level sentiment analysis and LIME-based explainability.
View case study
SmartGrowthAI is an end-to-end customer intelligence platform that combines churn risk prediction, demand forecasting, and NLP complaint analysis into an actionable React dashboard served via FastAPI.
View case study
Sentinel is a computer-vision system built to monitor personal protective equipment (PPE) compliance across active workspaces using YOLO, custom worker-equipment matching logic, and multi-camera support across web and desktop applications.
View case study
Sandesh is a multi-channel notification platform bringing push notifications, email, and SMS into a single delivery layer equipped with priority queues, retries, schedules, and dynamic templates.
View case study
Yojana Sahayak is a multilingual GovTech platform built to simplify scheme discovery, eligibility verification, document checklists, and application guidance using Playwright data pipelines, polyglot storage, and AI-assisted support.
View case study
Anvaya is an enterprise AI assistant built to unify 11 scattered organizational data sources into a single conversational search layer, using RAG with PostgreSQL-pgvector and MongoDB to deliver sub-3-second cited responses.
View case studyTell us what you're building, improving, or trying to figure out.