Hovhannes Tamoyan

Hovhannes Tamoyan

PhD at TU Darmstadt and Applied Research Intern at NVIDIA. I'm currently working on code and science reasoning.

Publications

↑ top

2026

Hovhannes Tamoyan, Sean Narenthiran, Erik Arakelyan, Mira Mezini, and Boris Ginsburg. "SHERLOC: Structured Diagnostic Localization for Code Repair Agents." arXiv preprint arXiv:2606.24820 (2026).
NVIDIA, including Hovhannes Tamoyan. "Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning." arXiv preprint arXiv:2606.15007 (2026).

2025

Hovhannes Tamoyan, Subhabrata Dutta, and Iryna Gurevych. "Factual Self-Awareness in Language Models: Representation, Robustness, and Scaling." arXiv preprint arXiv:2505.21399 (2025).
Hovhannes Tamoyan, Hendrik Schuff, and Iryna Gurevych. "Llm roleplay: Simulating human-chatbot interaction." Proceedings of the Third Workshop on Social Influence in Conversations (SICon 2025). 2025.

2023

Yu, Lili, et al. "Scaling autoregressive multi-modal models: Pretraining and instruction tuning (2023)." URL https://arxiv.org/abs/2309.02591.

2022

Hovhannes Tamoyan, Gayane Chilingaryan, Ani Tevosyan, et al. "BARTSmiles: Generative Masked Language Models for Molecular Representations. arXiv 2022." arXiv preprint arXiv:2211.16349.

2021

Hovhannes Tamoyan. "Bilingual Sentence Alignment Using Multilingual Language Models." Master's thesis, American University of Armenia (2021).

2020

Karen Hambardzumyan, Hovhannes Tamoyan, and Hrant Khachatrian. "YerevaNN's Systems for WMT20 Biomedical Translation Task: The Effect of Fixing Misaligned Sentence Pairs." Proceedings of the Fifth Conference on Machine Translation. 2020.

2019

Hovhannes Tamoyan, G. Zanyan, and H. Davtyan. "Low Energy Electron Beam Emittance Measurement at AREAL Accelerator." Armenian Journal of Physics 12 (2) (2019).

Projects

↑ top

Tools

clausius

Multi-cluster Slurm dashboard with AI agent integration via MCP. Monitor, explore, and manage GPU jobs across HPC clusters from a single browser tab — or let your AI coding agent do it through the built-in MCP server.

Urartu

ML pipeline framework with smart caching and remote execution. Chain actions into workflows, run anywhere — local machines to HPC clusters. YAML-configured with built-in experiment tracking and SLURM integration.

OrganizeNoc

Browser extension suite for researchers to manage academic papers library. Features metadata extraction, highlight processing, citation BibTeX export, and AI-powered querying.

Aim

Open-source experiment tracking tool designed with flexibility for machine learning workflows. Facilitates detailed monitoring and comparison of experiments.

tmynNLP

Natural language processing pipeline equipped with mainstream abstractions. Supports a diverse array of NLP tasks.

Teaching

Introduction to LLMs - Lecture (Notebooks)

Deep Learning - Lecture (Notebooks)

Experiment Tracking - Workshop

Essentials of experiment tracking in ML research, with an in-depth exploration of Aim and a practical demonstration integrating it into an NMT fine-tuning pipeline.

LxMLS Guide - Summer School

Lisbon Machine Learning Summer School lab guide with setup instructions, toolkit walkthroughs, and hands-on exercises.

PyTorch Optimization - Workshop

Strategies for efficient dataloader usage, parallel computation, operator fusing, and more to enhance PyTorch code performance.

Resume (download)

↑ top

Education

Ph.D, Computer Science, Technical University of Darmstadt

Research Areas: Natural Language Processing, Code Generation, Reasoning and Interpretability
Supervisor: Prof. Dr. Mira Mezini
2023 – present

M.Sc., Computer Science, American University of Armenia

Thesis: Bilingual Sentence Alignment in Low-Resource Settings
Supervisor: Karen Hambardzumyan
2019 – 2021

B.Sc., Nuclear Physics, Yerevan State University

Thesis: Electron Beam Emittance Measurement on the AREAL Linac
Supervisor: Prof. Dr. Vasili Tsakanov
2015 – 2019

Work Experience

Applied Research Intern, NVIDIA, Santa Clara, CA

  • Working on reasoning language models in the life science domain.
  • Contributed chemistry-domain training data to Nemotron 3 Ultra, an open-weight model (paper).
03/2026 – present

Applied Research Intern, NVIDIA, Yerevan, Armenia

  • Built SHERLOC, a code fault localizer, state of the art on SWE-Bench Verified and Lite, raising resolve rate at lower compute (paper).
05/2025 – 09/2025

Doctoral Researcher, Technical University of Darmstadt, Darmstadt, Germany

  • Published Factual Self-Awareness, showing LLMs encode whether they will recall a fact correctly (paper).
  • Published LLM Roleplay, a multi-agent system generating human-AI dialogues at scale (paper).
  • Built Urartu, a framework for one-command Slurm submission and reproducible runs.
04/2023 – 02/2026

Machine Learning Researcher, YerevaNN, Yerevan, Armenia

  • Contributed interpretability analysis to Chameleon, a multimodal model, with Meta FAIR (paper).
  • Explored transformer alternatives for text generation, with Meta FAIR: diffusion, mixture of experts, LM bootstrapping.
  • Trained BARTSmiles, a molecular representation model, with Meta FAIR (paper).
  • Ran cross-lingual zero-shot transfer experiments with prompting and adapters.
  • Solved bilingual sentence alignment in low-resource settings (Master’s Thesis).
  • Built a machine translation system for the WMT20 Biomedical tasks (paper).
09/2020 – 02/2023

Machine Learning Engineer, AimStack, Yerevan, Armenia

  • Integrated Aim experiment tracking with popular ML frameworks and MLOps tools.
  • Wrote 3 technical blog posts that grew community adoption.
08/2022 – 03/2024

Machine Learning Engineer, ZERØ, Yerevan, Armenia

  • Shipped a legal document classification system.
  • Built tmynNLP, the framework behind its multiclass classification pipeline.
11/2021 – 03/2022

Software Engineer, Birthright Armenia, Yerevan, Armenia

  • Led development of a large-scale content management and email system.
09/2019 – 07/2022

Physics Researcher, CANDLE SRI, Yerevan, Armenia

  • Measured electron beam emittance on the AREAL linac (paper).
  • Built tooling for linear accelerator control and beam diagnostics.
10/2018 – 11/2019

Web Developer, Floopen, Yerevan, Armenia

  • Built and maintained a general-purpose content management system.
12/2017 – 11/2018

Teaching Experience

Teaching Associate, Technical University of Darmstadt, Darmstadt, Germany

  • Taught Introduction to LLMs (09/2024 – 02/2025).
  • Mentored the experiment tracking and reproducibility workshop (10/2024).
  • Mentored the PyTorch optimization workshop (04/2023).
  • Supervised four M.Sc. theses in mechanistic interpretability.
04/2023 – 02/2026

Core Monitor, Instituto Superior Técnico, LxMLS 2023, Lisbon, Portugal

  • Wrote a step-by-step guide to the transformer architecture.
  • Led lab sessions for researchers, graduate students, and industry practitioners.
07/2023

Teaching Associate, American University of Armenia, Yerevan, Armenia

  • Deep Learning: taught NLP lectures and led problem sessions.
  • Data Structures and Algorithms: led problem sessions and wrote assignments.
09/2021 – 12/2021

Skills

Computer Science: Algorithm Design, Data Structures

Languages: C, Python, JavaScript

Machine Learning: PyTorch, TensorFlow, xFormers, FairSeq, AllenNLP, PyTorch Lightning

Database Systems: SQL, NoSQL, GraphQL