
👋 I'm Karan, a backend developer exploring Python, Node.js, and now machine learning. Keen on open source contributions to enhance my project management skills
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Obvix Labs
Data Scientist
June 14, 2026 – Present
claude-code-analysis
April 1, 2026 – Present
We read all 512K lines of Claude Code's accidentally exposed source. 82 docs, 15 diagrams, every subsystem mapped — from the hidden YOLO safety classifier to multi-agent swarms.
View Projectcontext_window_research
March 13, 2026 – Present
80,433-trial study of context-window sycophancy across 6 LLMs (4B–72B). Behavioral ratchet effect, correction injection mitigation, phase transition analysis. Code, data, and preprint included.
View Projectclaude-skills
February 12, 2026 – Present
Battle-tested Claude Code skills for hardening vibe-coded projects. Audit, clean, and ship AI-generated codebases with confidence.
View Projecthavoc-machine
December 27, 2025 – Present
Havoc Machine is a Synthetic Customer Chaos Lab that stress-tests customer support chatbots through AI-generated adversarial conversations. It automatically generates 1,000+ realistic chaos scenarios to expose policy vulnerabilities and refund leakage before impacting real customers.
View Projectclip-cast
October 1, 2025 – November 26, 2025
📺 YouTube Clone – MERN Stack A full-stack video streaming platform built with the MERN stack (MongoDB, Express, React, Node.js). Features include video upload, playback, likes, comments, and a modern UI inspired by YouTube.
View Projectobvix-wellness
September 21, 2025 – September 27, 2025
Obvix Wellness is a safety-first emotional AI chat companion combining retrieval-augmented generation, crisis detection, and locally served Fine-tuned models for grounded therapeutic conversations.
View Projectbodhi
June 28, 2025 – June 29, 2025
Dynamic knowledge base that auto-syncs your docs and FAQs from Google Drive, keeps company support/help pages current, and powers both a customer-facing support site and a real-time chatbot to answer questions instantly. Reduce support workload with consistent, always-updated answers based on your latest internal documentation.
View Projectsanskritayam
March 24, 2024 – April 2, 2024
"Sanskritayam" is a unique, experimental programming language inspired by the Sanskrit language. Developed for fun, following a viral meme, it integrates Sanskrit's grammatical precision into programming, featuring constructs like loops, conditions, data types, and OOP.
View ProjectShawtyGPT
February 21, 2024 – April 2, 2024
A fun project exploring the power of Google's Generative AI API to create a sassy, materialistic chatbot named "Lavanya". This project is meant to demonstrate the capabilities of large language models while acknowledging the potential sensitivities of the chosen persona.
View ProjectAFOptimizer
December 29, 2023 – Present
Enhance anime videos with AFOptimizer, a Python tool for removing static frames using Frame Difference , SSIM and Optical Flow methods. Ideal for streamlining viewing and editing.
View ProjectCultural Fit Analysis
The candidate's portfolio demonstrates a strong inclination towards innovative and experimental projects, often pushing boundaries in AI/ML. This suggests a fit for a culture that values research, development, and a proactive approach to technology. The personal projects, while diverse, are heavily concentrated in AI/ML and data science, aligning well with a Data Scientist role. The creation of a programming language and adversarial testing tools indicates a deep curiosity and a drive to understand systems thoroughly. However, the lack of team-based projects or contributions to open-source initiatives beyond personal projects makes it difficult to assess collaboration style.
Soft Skills & Operational Fit
The candidate's project descriptions suggest a proactive and experimental approach to problem-solving, indicating strong initiative and a willingness to explore novel solutions. The diversity of projects, from language creation to adversarial testing, points to adaptability and a broad interest in technical challenges. However, without psychometric test results, specific soft skills like logical reasoning, work attitude, stress handling, and team collaboration cannot be objectively assessed.