04 — About
Available for hire
I build things that hold up.
I'm Raghav, a software engineer. For about five years I built web platforms; these days I work mostly on AI and machine learning.
I started in frontend and full-stack roles at L&T and Country Delight: micro-frontends and a UI component library at one, a support chatbot, offer targeting, and recommendations at the other. More recently I benchmarked large language models for code generation at SnT, and I'm a full-stack developer at ArtAICare while finishing my Master's at the University of Luxembourg.
I'm based in Belval, Luxembourg, and open to engineering roles across the EU or remote.
Experience
2019 — nowFull-Stack Developer · ArtAICare
Oct 2025 – Present- Built the real-time collaborative canvas at the centre of the product — patient and therapist drawing on one surface during a live session, synced with the video and audio, with recording, replay, and version history.
- Rebuilt scheduling end to end: timezone-aware booking in two clicks, atomic guards against double-booking, and automated 24-hour and 1-hour reminders that open into the session room.
- Added TOTP two-factor auth and tightened access control on patient records and live-session data channels; took the whole app multilingual.
- Built an in-app AI help assistant (RAG) and supervised the Python image, recommendation, and music engines behind it.
- Ran the platform side — admin panel with role-based access and audit logging, CI/CD with auto-rollback, plus the earlier frontend revamp and performance work (lazy loading, code splitting, caching, query tuning).
Research Intern · SnT · TRUX Research Group, University of Luxembourg
Mar 2025 – Aug 2025- Built an evaluation set for benchmarking code-generation LLMs on top of CodeBenchGen, turning model output into executable tasks with test harnesses so generated code could be graded on whether it actually ran and passed.
- Measured correctness, runtime efficiency, and performance across static and dynamic settings, and tracked how stable models stayed as prompts and architectures changed.
- Set up the observability around it: per-task pass rates, failure modes, and run-to-run variance logged so the benchmark stayed reproducible and easy to debug.
- Worked on ML-based code authorship attribution for provenance and IP protection, alongside PhD researchers and toward published results.
Software Engineer · Country Delight · Bengaluru
Mar 2022 – Aug 2024- Built a rule-based chatbot for customer support, mapping common questions to guided answer flows and routing the edge cases on to a human.
- Built the customer-criteria component behind offers: the team could define a cohort by attributes and behaviour, then apply a promotion to exactly those users.
- Worked on the recommendation system that suggested what a customer might add next from their order history.
- Led the UI revamp of the customer-facing app, rebuilding it in Angular and Material-UI with a component-driven structure.
- Made the build smaller and the site faster through dynamic and lazy loading, image optimization, and performance profiling.
Software Engineer · L&T Technology Services · Mysore
Aug 2019 – Dec 2021- Built the ELCP dashboard for client ABB in Angular: a screen of operational data made of modular, plug-and-play widgets built from a range of charts and tables.
- Built a reusable UI component library in the spirit of Material UI, so other websites across the org could share the same components instead of rebuilding them.
- Documented every component in Storybook, with its states and variations, so other teams could adopt it without reading the source.
- Broke a large application into multiple micro-frontends with single-spa, and wired up the routing and communication between them so they ran as one app.
Education
University of Luxembourg
M.Sc. in Information & Computer Science
Sep 2024 – Present
National Institute of Engineering, Mysore
B.E. in Computer Science Engineering — 9.04 / 10
Aug 2015 – Jun 2019
Skills
LanguagesPython · C++ · JavaScript / TypeScript · SQL · Bash
AI / MLPyTorch · TensorFlow · scikit-learn · Keras · OpenCV
LLM / EvalCode-gen benchmarking · CodeBenchGen · eval-set design · evaluation pipelines · observability
DataNumPy · pandas · Matplotlib · Seaborn · Jupyter · FastAPI
FrontendReact · Angular · Next.js · Material-UI · HTML / CSS
DevOps & CloudGit · Docker · GitHub Actions · AWS (EC2/S3) · Linux
Want the short version on paper, or a conversation?