AI Engineer & Multi-Agent System Architect
Currently crafting AI magic at Simform's Innovation Lab. I turn LLMs into productive team members and build systems that actually work (most of the time). This portfolio was crafted with some serious vibe coding energy ✨
So here's the deal - I'm currently an AI Engineer at Simform's Innovation Lab, where I spend my days building multi-agent systems and convincing LLMs to behave like productive team members. Before this, I spent 3+ years at Tecblic turning machine learning dreams into actual working applications.
My specialty? Taking cutting-edge AI research and making it actually useful for real businesses. I've built everything from automated pitch generation systems (yes, AIs can now do sales better than most humans) to AI-powered mock interview platforms that probably judge candidates less harshly than I judge my own code.
When I'm not teaching machines to think, you'll find me diving deep into the latest GenAI trends, contributing to MCP servers, or explaining to yet another stakeholder why their AI assistant can't actually read their mind (yet).
My approach to AI development? "Learn. Implement. Iterate." It's served me well through prompt engineering marathons, API integration challenges, and those 3 AM moments when you finally get your multi-agent system to stop arguing with itself.
Fun fact: This entire portfolio was born from a late-night coding session fueled by questionable amounts of caffeine and some serious vibe coding energy. Several LLMs were mildly inconvenienced in the making of this website.
Years of AI Magic
APIs Built
Prompt Engineering
Here are some of the AI-powered systems I've built that are currently making life easier for real humans (and probably plotting world domination in their spare time):
Built a team of AI agents that can research prospects and generate pitches better than most sales teams. The CEO was so impressed, they're probably considering replacing half the sales department with bots.
Created an AI interviewer that's probably less biased than most human interviewers. It can analyze speech, evaluate candidates, and give feedback without needing coffee breaks.
Extended open-source PR Agent with custom lint fixes and bug finder tools. Now GitHub Copilot can find bugs faster than I can create them (which is saying something).
Built a system using pre-trained YOLO models to read receipts better than I can read my own handwriting. Features image enhancement with ESRGAN because blurry receipts are the bane of my existence.
Built a system that detects brand logos in real-time and overlays ads dynamically. Basically turned the world into one giant billboard (sorry about that).
Built an AI-powered medical assistant using pre-trained LLM models integrated with a Rasa chatbot. Users can input symptoms and get disease predictions - basically WebMD but actually useful.
Currently working in the R&D department, building cutting-edge AI solutions for C-suite executives. Basically, I'm the person who makes AI do the hard work while I drink coffee and pretend to understand what the models are thinking.
Spent 3 years turning machine learning research papers into actual working applications. Built everything from YOLO models that could spot brand logos to Flask APIs that survived production traffic.
Achieved 8.2/10 CGPA while learning the fundamentals of turning ideas into code. Also discovered my love for machine learning and my mild addiction to debugging.
Whether you've got a wild project idea, need someone to debug your "impossible" issue, or just want to chat about the latest JavaScript framework drama, I'm all ears!