Skip to content
View tim-ponomarev's full-sized avatar
1 follower · 0 following

Block or report tim-ponomarev

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
tim-ponomarev/README.md

Hi, I'm Tim

Applied ML Engineer working on LLM fine-tuning, production RAG, and reinforcement learning. Independent B2B contractor — open to long-term engagements (6+ months) or well-scoped projects.

Recent open-source

  • qlora-finetune — QLoRA fine-tuning of 7B models with reproducible training, eval, and merge-for-vLLM tooling
  • hybrid-rag — Production RAG pipeline: BM25 + dense + RRF + cross-encoder rerank + 3-axis LLM-as-Judge eval
  • decision-transformer-gtrxl — Decision Transformer with GTrXL backbone and World Model auxiliary head, plus a C++17 / AVX2 batched env via pybind11

Stack

LLMs — PyTorch · HuggingFace Transformers · PEFT · TRL · vLLM · LangChain · Qdrant · sentence-transformers ML — Reinforcement Learning · Decision Transformer · World Models · CatBoost · Weights & Biases Systems — C++17 · AVX2 SIMD · pybind11 · CUDA · Docker · Linux

Reach me

Email: timponomarev.dev@gmail.com

Pinned Loading

  1. decision-transformer-gtrxl decision-transformer-gtrxl Public

    Decision Transformer with GTrXL backbone and World Model auxiliary training, plus C++/AVX2 batched env

    Python 1

  2. hybrid-rag hybrid-rag Public

    Production-grade hybrid RAG: BM25 + dense retrieval + cross-encoder reranking + LLM-as-Judge

    Python

  3. qlora-finetune qlora-finetune Public

    QLoRA fine-tuning of 7B models on instruction datasets

    Python