<!-- Header banner -->

<p align="center"> <img src="https://capsule-render.vercel.app/api?type=waving&height=220&text=Bhavishya%20Pandit&fontSize=54&fontAlignY=38&fontColor=ffffff&desc=Turning%20AI%20into%20Enterprise%20Value%20%7C%20Google%20ML%20Expert%20%7C%20Top%20300%20Globally&descSize=17&descAlignY=74&animation=fadeIn&color=0:0B1221,100:0EA5E9&section=header" /> </p> <p align="center"> <a href="https://readme-typing-svg.demolab.com/demo/"> <img src="https://readme-typing-svg.demolab.com?font=Fira+Code&size=17&pause=700&duration=2600&center=true&vCenter=true&width=1000&lines=I+don%27t+just+build+AI+%E2%80%94+I+build+what+AI+needs+to+survive+itself;Production+ML+%2B+GenAI+that+ships%2C+scales%2C+and+moves+business+metrics;$XXM+in+enterprise+value+%7C+50M%2B+in+orbit+%7C+Speaker+-+MHA/IITs/NITs" /> </a> </p> <!-- Badges --> <p align="center"> <img src="https://komarev.com/ghpvc/?username=bhav09&style=for-the-badge" /> <img src="https://img.shields.io/github/followers/bhav09?label=Followers&style=for-the-badge" /> <img src="https://img.shields.io/badge/PyPI-Jhadoo-0EA5E9?style=for-the-badge&logo=pypi&logoColor=white" /> <img src="https://img.shields.io/badge/PyPI-Chowkidar-10B981?style=for-the-badge&logo=pypi&logoColor=white" /> </p> <p align="center"> <a href="#about">About</a> • <a href="#open-source--the-ai-infrastructure-layer">Open Source</a> • <a href="#impact-highlights">Impact</a> • <a href="#tech-stack">Tech</a> • <a href="#currently-exploring">Now</a> • <a href="#connect">Connect</a> </p>

About

<img align="right" alt="Night coding animation" src="https://raw.githubusercontent.com/AVS1508/AVS1508/master/assets/Night-Coding.gif" width="300" />

I turn AI into enterprise value. My impact spans across engineering, business, sales, community, content and public speaking.

Most people build AI for users. I also build AI for AI — the tools, guardrails, and infrastructure that keep production AI running efficiently.

Open source — the AI infrastructure layer

Most people ask: "How do I make AI faster?" I ask: "What happens when AI breaks itself — and how do we prevent it?"

These two tools come from that question.

🧹 Jhadoo — pip install jhadoo

The disk cleanup tool built for the speed of modern AI development.

ML engineers move fast. Conda environments, Docker images, stale branches, model checkpoints — they pile up silently until your machine crawls. Jhadoo sweeps it all up.

| Feature | What it does | |---|---| | 🌿 Smart Git & Docker | Finds stale branches and prunes Docker images older than 60 days | | 🛡️ Safety first | Dry-run mode shows you exactly what will be deleted before anything happens | | 📦 Archive mode | Not ready to delete? Move files to archive instead | | ↩️ Instant undo | Deleted something by mistake? Restore it immediately | | ⏰ Built-in scheduler | Set it to 2 AM. Wake up to a clean, fast machine | | 📊 Dashboard | Track storage trends with progress bars and live notifications |

Works on macOS, Windows, and Linux.

• 📦 4,000+ downloads
• 🧹 Average savings: 18 GB per developer
• 🏆 One ML engineer cleared 37 GB in seconds

🛡️ Chowkidar — pip install chowkidar

The Dependabot for your AI models. Because AI models expire — and nobody tells you until production crashes.

In 2024 alone, OpenAI retired 7 model variants. If you use gemini-1.5-flash, it might die on June 6. Without Chowkidar, you find out when the app breaks on June 7. With Chowkidar, you get a heads-up in March and fix it in 10 minutes.

Dependabot tells you when your Python libraries are old. Chowkidar tells you when your AI models are about to kill your product.

| Capability | Details | |---|---| | 🔍 Scans | .env, YAML, TOML, JSON, Python, JS, TS files for every model string | | 🔄 Syncs | Official sunset schedules from OpenAI, Anthropic, Google, Mistral | | 🚨 Warns | Desktop + terminal alerts at 90, 30, and 7 days before deprecation | | 💰 Fixes | One-click replacements with cost savings and capability diff shown upfront | | 🚦 CI/CD gate | chowkidar gate in GitHub Actions — fails the build if an expired model ships | | 🤖 Zero-config IDE rules | Auto-generates rules for Cursor, Claude Code, VS Code, Windsurf | | 🖥️ MCP server | Your IDE can query model status in real-time | | 👁️ Background daemon | Watches your files 24/7 on macOS, Linux, Windows | | 🔒 Local-first | Ollama + Gemma for smart logic. Your API keys never leave your machine |

Featured on Google Cloud — Community (Medium).

• 📦 2,000+ downloads in weeks
• ⚡ Turned a potential "2 AM production crash" into a "5-minute planned update"
• 💸 Switching models via Chowkidar can save up to 96% on token costs

Impact highlights

<table> <tr> <th align="left">Area</th> <th align="left">What I've done</th> <th align="left">Proof / outcomes</th> </tr> <tr> <td><b>Engineering</b></td> <td>Owned the full AI lifecycle: discovery → architecture → governance → production</td> <td>Rebuilt key workflows end-to-end; delivered up to <b>60% productivity gains</b></td> </tr> <tr> <td><b>Enterprise</b></td> <td>Built AI accelerators used by GTM teams to close deals faster</td> <td>Contributed to <b>$XX Million</b> in enterprise deal value</td> </tr> <tr> <td><b>Public sector</b></td> <td>Built an internal AI productivity platform for Government of Uttarakhand</td> <td>Deployed to <b>5,000+ officers</b> incl. IAS/IPS; cut analysis time by <b>98%</b></td> </tr> <tr> <td><b>Open source</b></td> <td>Jhadoo (disk hygiene for ML devs) + Chowkidar (model lifecycle watchdog)</td> <td><b>6,000+ downloads</b> combined; featured on Google Cloud — Community</td> </tr> <tr> <td><b>Tourism</b></td> <td>AI-powered offline experiences with Saudi Tourism Authority</td> <td>Reached <b>10,000+ people</b></td> </tr> <tr> <td><b>Thought leadership</b></td> <td>LinkedIn Top Voice in AI · Google Developer Expert (GDE) in AI · Substack</td> <td><b>50M+ content views</b> · <b>250K+ weekly readers</b></td> </tr> <tr> <td><b>Speaking</b></td> <td>100+ talks at MHA Govt. of India, IITs, NITs, IIITs</td> <td>Live audience of <b>25,000+</b></td> </tr> <tr> <td><b>Hackathons</b></td> <td>2× GenAI hackathon winner — RAG pipelines + multi-agent orchestration</td> <td>Award-winning prototypes in security and LLM observability</td> </tr> </table>

Connect

<p align="center"> <a href="https://www.linkedin.com/in/bhavishya-pandit-68a4a018a/"> <img src="https://img.shields.io/badge/LinkedIn-Connect-0A66C2?style=for-the-badge&logo=linkedin&logoColor=white" /> </a> <a href="https://twitter.com/BhavishyaP9"> <img src="https://img.shields.io/badge/X-Follow-111111?style=for-the-badge&logo=x&logoColor=white" /> </a> <a href="https://bhavishyapandit9.substack.com"> <img src="https://img.shields.io/badge/Substack-Subscribe-FF6719?style=for-the-badge&logo=substack&logoColor=white" /> </a> </p> <p align="center"> <i>If you want to scale AI adoption, automate high-impact workflows, or get measurable results from GenAI + ML — let's connect.</i> <br/> <i>If you need a speaker, advisor, or someone who can see the AI opportunity before your competitors do — hit me up.</i> </p> <p align="center"> <img src="https://capsule-render.vercel.app/api?type=waving&height=100&color=0:0EA5E9,100:0B1221&section=footer" /> </p>