Skip to content
View vignesh-nagarajan-vn's full-sized avatar
🎯
Focusing
🎯
Focusing
9 followers · 9 following

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull SharkAchievement: QuickdrawAchievement: YOLO

Achievements

Achievement: Pair Extraordinairex2Achievement: Pull SharkAchievement: QuickdrawAchievement: YOLO

Highlights

  • Pro

Block or report vignesh-nagarajan-vn

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
vignesh-nagarajan-vn/README.md

Vignesh Nagarajan

Texas A&M Engineering · Computer Science × Econometrics
Machine Learning | Game Theory & Decision Science | Cognitive Modeling

Website LinkedIn Email

Technical Roles

  • Technology & Implementation Consulting Extern — Deloitte, Tempe Office
  • Founding Team Member — ScienceFair.io / Qurios Ventures (Acquired)
  • Data Analyst Intern — STEM·E Career Development Program
  • Software Development Intern — Jetson ($3M Series A Raise)
  • Data Science & Machine Learning Intern — Drippy

Selected Projects

  • Norn — Variant-interpretation copilot that drafts ACMG/AMP evidence for a human curator: paste a genetic variant (HGVS, rsID, or locus) and Norn pulls from public genomics databases, adjudicates each criterion with a two-pass Claude reviewer, then computes the classification in code. Built at Claude Life Sciences Hackathon (1 of 500 selected from 6,000+ applicants).

  • SchoolPulse AI — Edge-AI suite with statistically-gated leak detection (confidence-interval slope analysis on simulated sensor data), an EfficientNet-based waste-sorting classifier (96% accuracy, deployed via TFLite), and a RAG-driven operations agent, all unified via a dashboard tracking a school's water, food, and energy footprint. Finalist at USAII Global AI Hackathon.

  • Ordinance — Full-stack AI backlayer for Congressional offices, running a keyword-scored RAG pipeline over a curated federal corpus (EPA rules, OMB circulars, CISA directives) that returns citation-bracketed answers anchored to CFR/USC references, with all inference proxied through serverless API routes so credentials never reach the browser.

Currently building REFLEX (Reflexive Equilibrium Fixed-point Learning for endogenous financial markets).

Research Papers

  • Nagarajan, V. (2026). MIRROR: Multimodal Intelligent Radiology Reasoning and Observation Reporter. arXiv Computer Science, Computer Vision and Pattern Recognition (cs.CV) — Pending Preprint
  • Nagarajan, V. (2026). CANVAS: Captioning Art with Narrative Visual-Audio AI Systems. arXiv Computer Science, Human-Computer Interaction (cs.HC). https://doi.org/10.48550/arxiv.2606.09846
  • Nagarajan, V. (2024); Mittal, A. (2024). PEBSA: Public Economic Behavior Sentiment Analysis. International Journal of Engineering and Computer Science (IJECS). https://doi.org/10.18535/ijecs/v13i12.4950

Open-Source Contributions

  • yfinance — Three upstream fixes to the most widely used Python library for Yahoo Finance market data: (1) fixed a 4-year-old bug where 30-min price requests raised misleading unavailability errors from an internal interval substitution (PR #2900 for Issue #1029); (2) repaired pandas 3 test-suite breakage from zoneinfo vs pytz timezone comparisons (PR #2901); (3) stopped error messages from falsely claiming tickers are "possibly delisted" (PR #2903 for 5 Issues incl. #2902).

  • qlib — Two contributions to Microsoft's AI-quant investment platform: (1) fixed two long-standing FileInstrumentStorage bugs that were corrupting qlib's instrument files, adding regression tests plus a new self-contained class and 7 new functions (PR #2302 for Issue #1720); (2) aligned the pre-commit black formatting hook with the version CI enforces (PR #2303).

Relevant Awards

  • Mu Alpha Theta National Scholar — 1 of 27 Scholars, $4K Award, First-ever Arizona Winner
  • Congressional App Challenge Winner — Presented Chronos at the US Capitol
  • National Cyber Scholar — SANS-GIAC Cybersecurity Scholarship Recipient, $3.4K Award
  • CyberStart America Hackathon Winner — Top ~2.8% of 35K+ Competitors
  • AMC / AreteLabs Mini-Math Madness — 2× 1st Place (National Champions) & 1x 2nd Place

Always open to interesting problems and collaborations.
vigneshn26@tamu.edu

Pinned Loading

  1. Norn Norn Public

    Drafts ACMG/AMP variant classifications for a human curator: gathers genomics evidence, adjudicates each criterion with Claude, computes the label in code via ClinGen points, and shows every source.

    TypeScript 1

  2. SchoolPulse-AI SchoolPulse-AI Public

    Suite of edge-AI tools and a web dashboard that help schools cut water, food, and energy waste by detecting leaks, sorting compost, and visualizing consumption in one conclusive dashboard.

    Jupyter Notebook 3 1

  3. REFLEX REFLEX Public

    ML for endogenous OTC bond markets, where a dealer's policy reshapes future flow. Reframes learning as finding a fixed-point equilibrium stable under self-induced market adaptation.

    Python 1

  4. Ordinance Ordinance Public

    RAG-powered operating system for Congressional offices that turns legislation, ethics rules, and appropriations into "Actionable Insights" (AI), answering natural-language queries with cited sources.

    TypeScript 1

  5. MIRROR MIRROR Public

    Explainable radiology AI that reads chest X-ray, brain MRI, and head CT scans, then chains classification, evidence localization, and language generation into a traceable clinician-style report.

    Python 1

  6. Chronos Chronos Public

    Forked from Raghavsk24/Chronos

    Automatically schedules meetings for large groups on Google Calendar using availability analysis and multi-factor slot scoring.

    TypeScript