Karya
An offline AI education platform combining quantized on-device LLM inference, multilingual RAG across 22 Indian languages + English, document OCR, and interactive whiteboard tools.
AI Engineer
OpenSource Project Creator & Maintainer · Zero-API Developer CLI
Built an open-source, zero-API developer CLI that guides engineers through structured, end-to-end software development phases using any AI model (ChatGPT, Claude, Cursor, Gemini, local LLMs). Designed with a local-first memory architecture (.buildwithai/context.json), prerequisite validation, automated clipboard synchronization, and 6 production templates.
Open Source Contributor · Serverless Hybrid Vector Database
Engineered core vector database algorithms, in-memory SPFresh dynamic partitioning, FastMCP server for GraphRAG, AWS pre-flight diagnostic CLI (dynavec doctor), SHA-256 ingestion deduplication, PEP 561 typing compliance, and distributed multi-OS regression suites across 11 merged PRs.
An offline AI education platform combining quantized on-device LLM inference, multilingual RAG across 22 Indian languages + English, document OCR, and interactive whiteboard tools.
An AI-powered video comment intelligence platform that automatically clusters YouTube comments into interactive RAPTOR topic trees, audience sentiment analytics, toxicity audits, and conversational insights.
A browser-based pair programming tool featuring real-time collaborative code editing synchronized via WebSockets, multi-language execution across 7 languages, and integrated chat.
A full-stack incident management system with automated SLA enforcement, Celery asynchronous alerts, RBAC, audit trails, and reliability analytics.
An autonomous AI coding agent that plans, modifies, tests, and independently verifies real-world codebases with sandboxed isolation and automatic rollback.
A CLI toolkit for benchmarking AI coding agents across 40+ programming tasks using deterministic rule-based scoring and full execution trace logging.
An intelligent proxy gateway that dynamically routes LLM queries based on prompt complexity—directing simpler queries to lower-cost models while reserving premium models for complex tasks.
Full-featured architectural sketching canvas powered by Excalidraw for designing distributed systems and algorithms.
Custom productivity suite featuring Pomodoro Deep Work with floating PiP capsule, atomic habits, and activity heatmap.
Distraction-free personal notepad with multi-notebook organization, zero typing lag, tag filtering, and auto-saving.
Dedicated quick-capture markdown scratchpad with multi-sheet tabs and instant export. Accessible while reading technical blogs.
Spotlight command center for lightning-fast site navigation, deep search, focus timers, and quick developer tools (Ctrl+K).
Deep dives into Transformer mechanics, quantization, local inference with Llama.cpp, and vector embeddings.
Scalability blueprints, cache invalidation strategies, database replication, and resilient microservice architectures.
Core algorithmic paradigms, dynamic programming optimizations, graph traversals, and time-complexity proofs.