Aspiring AI/ML and Data Science professional. I care about building things that actually work, not just things that look impressive.
I'm Impana T, deeply interested in machine learning, artificial intelligence, and data science. I want to build systems that solve real problems — not just impressive demos.
Whether it's wrangling messy datasets, fine-tuning neural networks, or telling a clear story with data, I bring both rigour and curiosity to the work.
Currently exploring roles in AI/ML Engineering, Data Science, and MLOps.
Here are some technologies I have been working with:
Outside of work, I'm nerdy about dinosaurs, love volleyball, and watch way too many crime documentaries.
Explainable AI for protein hydration sites with quantum ML for uncertain cases. Combines Biopython and LLMs to deliver interpretable predictions.
Visualises orbital debris density using real TLE data. Highlights high-risk zones with an interactive heatmap overlay across orbital shells.
Agentic pipeline that crawls, audits, and rewrites page content for SEO gains autonomously. Powered by a tool-calling LLM with a structured feedback loop.
CNN classifier detecting 15+ plant diseases from leaf images. Achieved 91% validation accuracy on a custom-augmented dataset with real-world noise.
LSTM network predicting short-term equity prices. Trained on 5 years of historical OHLCV data with technical indicators as engineered features.
Fine-tuned transformer model that condenses radiology reports into plain-language summaries, reducing reading time by ~70% in user testing.