M.Sc. Data Science & AI · Saarland University

Machine learning, end to end.

I work across NLP, computer vision and applied ML, taking problems from dataset construction through model benchmarking to a documented, reproducible result. First author of a peer-reviewed computer vision paper (Sādhanā, Springer 2025). Python is my daily driver, with the OpenAI and Anthropic APIs for LLM tooling and agentic workflows.

Open to Werkstudent roles · from Aug 2026 · Saarbrücken / remote Now: reproducing MolCA (EMNLP 2023), graph-augmented LLMs
kp_00 · 0.98 kp_06 · 0.94
Peer-reviewed

Real-time yoga pose estimation with a hybrid MoveNet architecture

Sādhanā · Indian Academy of Sciences, Springer · 2025 · First author

Built during a computer vision research internship at NIT Puducherry. The architecture splits deliberately: MoveNet Thunder generates high-fidelity keypoints for training, while MoveNet Lightning handles inference at 30+ FPS on consumer hardware: accuracy where the model learns, speed where the user waits. I owned the dataset end to end, consolidating and curating 3,435 images across 11 asanas from Kaggle and Yoga-82 sources, then trained the keypoint-embedding classifier, ran the misclassification analysis and carried the paper through two rounds of peer review.

99%
classification accuracy
11
asana classes
3,435
images curated
2-stage
train / infer split
Selected work
In progress · LLMs × Graphs

MolCA reproduction: can an LLM read molecular graphs?

Fall 2026 graduate seminar on transformer language models and graph integration with Prof. Dietrich Klakow, Saarland University. My first-choice paper: MolCA (Liu et al., EMNLP 2023), which gives a text-trained LLM 2D graph perception by wiring a molecular graph encoder into Galactica through a Q-Former cross-modal projector, with a LoRA adapter for downstream fine-tuning. The coding task (roughly 60 hours) is a critical verification, not a demo: reproduce a downstream result, run the ablation behind the paper's central claim (does the graph modality actually help, or is the LLM doing the work alone?) and test generalization on data the authors didn't use. Culminates in a 10-page thesis-style report and a 35-minute final talk in September.

PyTorch · Hugging Face Transformers · GNN graph encoder · Q-Former (BLIP-2-style) · LoRA · ablation design
Applied ML · Consulting

Lingomatch, AI document triage for a Gov/Justice-Tech platform

Student consultant (AI/Data Science) via the UdS Innovation Center's Triathlon startup programme: I built the ML backend of an AI-driven document-triage feasibility study for Lingomatch, turning photos of foreign-language documents into usable German gist translations for public authorities. I built and benchmarked the Arabic→German pipeline: Surya OCR evaluated on KITAB-Bench with CER/WER and confidence calibration; evaluated PaddleOCR as an independent second engine for disagreement-based flagging and documented its failure mode on Arabic due to engine-asymmetry; and benchmarked Opus-MT against NLLB-1.3B, matching translation quality (BLEU 43.8 vs 42.4) at a ninth of the memory, after catching the licence issue that ruled NLLB out for production.

Delivered the live demo in the final case-study presentation to the company's CEO.

Python · Hugging Face Transformers · Surya OCR · PaddleOCR · Opus-MT · NLLB · BLEU / CER evaluation
NLP · Pipeline

DODI: deception scoring for Terms-of-Service documents

DODI (Digital Ownership Deception Index) quantifies the gap between "Buy now" marketing and "you're only licensing this" fine print in consumer ToS documents. A deterministic, fully reproducible scorer (Flesch–Kincaid readability, licence-vs-ownership term ratio, red-flag clause detection, weighted 25/50/25) validated against ToS;DR human expert grades (Spearman ρ = 0.54, n = 12) and applied as a temporal study: 10 platforms, four Wayback Machine snapshots each (2015–2024). The index passes its face-validity test: GOG, the DRM-free "you own your games" store, scores lowest of all ten platforms.

Python · textstat · pandas · SciPy · ToS;DR validation · Wayback Machine corpus
Data analysis · Economics

Shadow Correction: piracy as a market signal

Quantitative study (team of 3) testing whether piracy behaves as a market-correction signal, structured as three testable barriers. The primary parameter: streaming-market fragmentation, measured with the Herfindahl–Hirschman Index on Parrot Analytics demand-share data. The supporting parameters were designed as falsification tests: piracy-intent indices built from Google Trends (pytrends) aligned with World Bank PPP-adjusted hourly wages across 90+ countries, where the statistically-zero economic correlation (Spearman ρ = −0.001, p = 0.99) is itself the finding, isolating fragmentation rather than affordability as the driver; plus event-study analysis of piracy spikes after content delistings.

Python · pandas · pytrends · HHI · Parrot Analytics / World Bank data · event-study design
Tooling & stack

Build with

PythonPyTorchTensorFlow/Keras Hugging Face Transformersscikit-learnspaCy pandasNumPySQL RGitLaTeX

LLM work, daily

OpenAI APIAnthropic API prompt designembedding-based retrieval LLM output evaluationagentic workflows Claude CodeCursor

How I work

I benchmark before I recommend: models get tested on realistic inputs, and failure modes go in the report next to the wins; a translation model that degrades on long legal text is a finding, not a footnote. I take deployment constraints seriously from day one: model licences, GDPR, EU AI Act obligations and what actually fits in 8 GB of VRAM. Everything ships version-controlled and documented so it survives handover. Where I'm still building depth I say so and pick it up fast: Docker, CI/CD and cloud deployment were on that list until I shipped DODI's live scorer as a Dockerized FastAPI service with CI on every push.

EN: C1 fluent · DE: A2, in active study toward B1 · Tamil: native · Hindi: conversational
Path
2026–now
MolCA reproduction, graduate seminar with Prof. Dietrich Klakow, Saarland University. Critically verifying a graph-augmented LLM paper (EMNLP 2023).
2026
Student consultant (AI/Data Science), lingomatch GmbH via the UdS Triathlon case study programme. Built and benchmarked the Arabic→German document pipeline; live demo to the CEO.
2025
First-author paper published in Sādhanā (Indian Academy of Sciences, Springer): real-time yoga pose estimation with a hybrid MoveNet architecture.
2024
Started M.Sc. Data Science & AI, Saarland University. Earlier that year: computer vision research internship at NIT Puducherry, where the paper's work was done.
2023
Web development intern, Zoho Corporation, Chennai. Full-stack library app with a content-based book recommender.
2020–2024
B.Tech Artificial Intelligence & Data Science, Bannari Amman Institute of Technology (CGPA 8.65/10). Thesis: LipNet-based visual speech recognition for hearing-impaired users.