I will fine tune llms using lora or qlora for your custom use case

Drona AI

India

I speak English

AI and ML Engineer for LLM Fine Tuning RAG and Data Science

AI/ML developer with hands-on production experience in LLM fine-tuning (QLoRA/PEFT), RAG systems (ChromaDB, MongoDB Atlas vector search, agentic memory), and data engineering (PySpark, Airflow, Delta ...
Read More
About this Gig

I fine-tune open-source LLMs (Llama, Mistral, Phi-3, etc.) on your custom dataset using LoRA/QLoRA parameter-efficient methods that don't require retraining the full model.

What I handle:

  • Dataset formatting and validation for your task (classification, generation, instruction-following)
  • LoRA rank/alpha configuration suited to your data size
  • Training with quantization (bitsandbytes NF4) to keep compute costs down
  • Evaluation against a held-out set, with clear before/after metrics

I won't oversell what fine-tuning can fix if your problem is better solved with prompt engineering or RAG instead, I'll tell you before we start, not after you've paid for training.

Send your dataset format and target task in the requirements, and I'll confirm scope before starting.

Read More
fine tune llms using lora or qlora for your custom use case
Full Screen
fine tune llms using lora or qlora for your custom use case

Programming Language:

Python

•

Javascript

•

Tensorflow

AI Model Frameworks & Tools:

PyTorch

•

NVIDIA CUDA & cuDNN

•

MLflow

Data Type:

Text

•

Images

•

Multimodal

AI Engine:

Stable diffusion

•

TensorFlow

•

Bert

•

Llama

•

Langchain

Related tags