Hi, I’m Deepesh Krishnan 👋

🚀 Creating DataStruk

I’m the creator of DataStruk CodeSense, working at InnerLogic and deeply passionate about coding, emerging technologies, and constantly experimenting with new ideas.

TL;DR: I’m that developer who can’t stop building things and asking “what if we could make this better?”

💻 My Journey

Always Experimenting

I’m the kind of developer who can’t resist trying new technologies, frameworks, and approaches. Whether it’s the latest AI model, a promising new programming language, or an innovative developer tool - I’m probably already testing it in a side project.

My GitHub is basically a graveyard of experiments, half-finished prototypes, and “what if…” projects. Some die quietly, others evolve into something useful. DataStruk CodeSense is one that refused to stay dead.

Working at InnerLogic

Currently building solutions at InnerLogic, where I get to work on challenging problems and explore how technology can solve real-world issues. It’s the perfect environment for someone who loves to push boundaries and isn’t afraid to break things in the process.

The DataStruk CodeSense Story

DataStruk CodeSense emerged from a simple frustration: every time I wanted to get AI help with my code, I had to manually copy-paste files, explain context, and repeat the same process across different AI assistants.

I thought: “What if any AI could instantly understand any codebase?”

That late-night thought led me down a rabbit hole of:

  • Discovering the Model Context Protocol (MCP)
  • Realizing we could build something universal
  • Countless weekends prototyping and rebuilding
  • Testing with every AI assistant I could get my hands on

🔧 What Drives Me

The “Why Not?” Philosophy

Most people ask “Why?” when they see something new. I ask “Why not?”

  • Why not make AI universally understand code?
  • Why not build bridges instead of walls between tools?
  • Why not experiment with that crazy idea that just popped into my head?

Open Source Believer

DataStruk CodeSense is open source because the best tools are built by communities, not corporations. I’ve learned more from GitHub repositories than any textbook, and I want to pay that forward.

Quality Over Speed

In a world of “move fast and break things,” I prefer “move thoughtfully and build things right.” DataStruk CodeSense took longer to build because I wanted it to be something I’d be proud to use every day.

🎯 Beyond Code

When I’m not coding (rare, but it happens):

  • 🤖 AI Tinkering: Exploring new models, fine-tuning experiments
  • 📚 Tech Reading: Always have 5+ browser tabs open with interesting papers
  • 🔬 Side Projects: Usually working on 2-3 “wouldn’t it be cool if…” ideas
  • 🌱 Learning: Currently diving deep into systems programming and protocol design

🚀 What’s Next?

DataStruk CodeSense is just the beginning. I’m working on:

  • More MCP Tools: Expanding the intelligence suite
  • Better Integrations: Making it work seamlessly with more AI assistants
  • Community Features: Tools for teams and collaborative development
  • Secret Projects: A few things I can’t talk about yet 😉

The goal? Make every developer’s interaction with AI more intelligent, efficient, and delightful.

“The best way to predict the future is to build it.” - That’s my motto, and DataStruk CodeSense is my latest attempt at building a better tomorrow for developers.

🎯 Philosophy

Technology Should Adapt to You

I believe developers shouldn’t be forced to change their favorite tools. Whether you prefer Claude, ChatGPT, Groq, or local models - your AI assistant should work seamlessly with your codebase.

Standards Over Silos

Rather than building another proprietary tool, I chose to embrace MCP - an open protocol that benefits the entire developer community. When everyone wins, innovation accelerates.

Experimentation Drives Innovation

Most of my best ideas come from late-night coding sessions, trying “what if” scenarios, and combining technologies in unexpected ways. DataStruk itself started as one of these experiments.

🛠️ Tech I’m Excited About

Languages: Rust (for performance), TypeScript (for everything else), Python (for AI experiments)

Current Obsessions:

  • Model Context Protocol - The future of AI tool integration
  • Local AI Models - Privacy-first intelligence
  • Rust in Production - Building reliable, fast systems
  • Developer Experience - Making complex things simple

Always Learning: Whatever’s newest in AI, developer tooling, or system design

🔬 What I’m Building

Beyond DataStruk, I’m constantly working on experiments:

  • AI-powered developer tools that actually understand context
  • Local-first applications that respect privacy
  • Developer productivity hacks that integrate seamlessly into workflows
  • Open-source contributions to projects I believe in

💡 The DataStruk Vision

Every developer should have an AI assistant that truly understands their codebase.

This isn’t just about better code analysis - it’s about democratizing access to advanced development insights. Whether you’re a solo developer or part of a large team, whether you’re using cutting-edge AI or prefer local models, you should have the same level of intelligent code understanding.

🌍 Let’s Connect

I love connecting with fellow developers, especially those who share a passion for experimentation and pushing technological boundaries.

🔮 What’s Next?

I’m always experimenting with new ideas. Some current areas of exploration:

  • Multi-modal code intelligence - Understanding code through multiple AI modalities
  • Real-time collaboration tools - Making remote development feel local
  • Privacy-first AI development - Powerful tools that keep your code local
  • Next-generation developer workflows - Reimagining how we write and maintain code

“The best way to predict the future is to build it. The second best way is to experiment relentlessly until you understand what’s possible.”

- Deepesh Krishnan

Want to collaborate on something interesting? I’m always open to discussing new ideas, potential collaborations, or just geeking out about the latest in tech.