hello world, i'm
Rakshithaa
An engineer with a passion for exploring what AI can really do. Currently a research assistant at the Reading + Learning Lab, building AI that actually teaches.
An engineer with a passion for exploring what AI can really do. Currently a research assistant at the Reading + Learning Lab, building AI that actually teaches.
I've been really interested in this intersection of AI and how people actually learn. Not just building models that perform well on benchmarks, but figuring out how to make AI that genuinely helps someone think better. What does it look like when a system scaffolds your reasoning instead of just giving you the answer? That's the kind of stuff I think about a lot.
Lately I've been going down this rabbit hole of brain-computer interfaces and how the brain processes learning, and I kind of never came back. Mathis (2026) in Nature Neuroscience talks about "adaptive intelligence," basically how biological brains adjust on the fly in ways AI still can't. DeepMind put out a cognitive taxonomy paper that flagged five out of ten cognitive abilities (learning, metacognition, attention, executive functions, social cognition) as not even having good AI benchmarks yet. And NeuroChat (2025, ACM CUI) used EEG-based engagement tracking with an AI tutor. Students felt more engaged but didn't actually learn more. That gap between feeling like you're learning and actually learning really stuck with me.
Neuroscience is where I want to go next. I think if you actually want to build AI that works well with people, you have to understand how people think first. Not just the CS side, the brain side too. How do we take what we know about how the brain learns and use that to design better AI systems? That feels like the missing piece that nobody's really connecting yet, and I want to be the one working on it.
When I'm not in the lab I'm probably at the gym (strength training has become a whole thing), scrolling through my ever-growing collection of photos of my dog Yuuki back in India for a serotonin boost, or watching anime that makes me feel too many things. Frieren, Poco's Udon World, Naruto. I'll read a paper about brain-computer interfaces for fun and then go home and play Stardew Valley for four hours. I like a lot of different things and I don't think that needs an explanation.
My thesis explores whether different LLM tutor personas (a Socratic mentor, a debugging peer, and a vanilla control) produce meaningfully different learning outcomes in intro Python. It's a between-subjects RCT with diagnostic routing and multi-layer guardrails to keep the AI from just giving answers away.
Mentor (pump → hint → directive → assertion), Peer (collaborative debugging), Control (vanilla LLM).
Pre-test as diagnostic across 4 Python chapters. Weak chapters trigger tutoring + mastery checks.
Two-layer: prompt constraints + output filtering. Prevents answer leakage, keeps dialogue natural.
Designing an LLM-based adaptive tutoring system for introductory programming with 3 persona-driven experimental conditions, RAG-based content retrieval, and multi-layer guardrails. Built the full-stack application integrating Qualtrics survey flow with participant routing, session management, and interaction logging. Heading into a 150+ participant RCT on Prolific. Also developed an AI chatbot for correcting AI misconceptions (GPT-4.1, AWS Lambda) with a between-subjects experiment (N=200) across 3 conditions.
Publications: Psychonomic Society 2025 (poster) • ST&D 2026 (spoken presentation)
TA for EE3005 in the ECE department, supporting students through coursework, labs, and office hours.
Built an NLP + ML pipeline for multi-label Surgical Site Infection classification across 8,016 surgical encounters. Engineered 40+ polarity-aware NLP features from 6 clinical note types using BioMedICUS and a 172-concept UMLS lexicon. Trained XGBoost, CatBoost, and LightGBM with Optuna optimization, achieving F1 scores of 0.85–0.94 on a <3% prevalence dataset.
Built NLP pipelines for clinical text classification at UMN Medical School. The challenge was extracting structured labels from messy, unstructured clinical notes where doctors write in shorthand and abbreviations.
Key finding: Feature engineering on clinical text (n-grams, negation detection) mattered more than model complexity. XGBoost + Optuna beat deep learning approaches while being interpretable, which matters in clinical settings.
Replaced VGG19 with ResNet34 for neural style transfer and integrated Depth Anything V2 for spatial coherence. Multi-level AdaIN across five encoder levels with a U-Net decoder. Trained on COCO + WikiArt.
Key finding: ResNet needs higher style weighting than VGG (1.2/25.0 vs 1.0/10.0). Training from scratch beat fine-tuning pretrained weights. Main limitation: depth is injected at inference, not learned end-to-end, so the network never truly learns depth-specific features.
Built a system where an LLM reasons about spatial tasks, YOLO detects objects, ViT classifies them, and 3D RRT* plans collision-free paths. The LLM acts as a high-level planner while traditional algorithms handle the low-level execution.
Key finding: LLMs are surprisingly good at spatial reasoning when given structured scene descriptions, but hallucinate object positions without visual grounding. The pipeline approach (LLM + vision + planning) worked better than end-to-end.
Combined DDPG reinforcement learning with Spiking Neural Networks for energy-efficient robotic arm grasping. The idea was to see if biologically inspired spiking networks could match traditional deep RL while being more computationally efficient.
Key finding: SNNs showed promise for sparse reward settings but struggled with fine-grained motor control. Learned that the gap between biological and artificial neural computation is real and interesting.
Faster R-CNN with custom image enhancement for identifying marine species in degraded underwater imagery. Underwater images are noisy, color-shifted, and low-contrast, so preprocessing was half the battle.
Key finding: Domain-specific image enhancement (color correction, contrast normalization) improved detection accuracy more than switching to a bigger model. A reminder that data quality > model size.
$ cat beyond_the_code.txt
[LOADING...]
🐶 YUUKI 🐶
SPECIES: golden labrador
STATUS: professional couch thief
HOBBY: blanket hoarding
CUTENESS: ██████████ MAX
NAPS/DAY: ██████████ YES
best girl in all of india & minnesota 💖
xX_n4rut0_f4n_Xx: believe it!! great site 🔥
stardew_girl_22: omg the konami code is genius
soma_survivor: WAP = what am i? the real horror is philosophy.
frieren_enjoyer: this page heals my soul
gym_bro_99: calisthenics gang rise up 💪