Tanveer A.

Key Skills

Software

No software listed

Top Subject Matter

Medical LLM fine-tuning and domain adaptation

Top Data Types

Image

Top Task Types

Fine-tuning

Freelancer Overview

Internee, AI development with medical LLM (QLoRA/PEFT) fine-tuning project work. Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Unsloth and Google Colab. Education includes Matriculation (Science), G.H.S.S (2020) and Intermediate (FSC), Ghazali Science College (2022). AI-training focus includes data types such as Computer Code, Programming, and Image and labeling workflows including Fine-tuning, Prompt-Response Writing, and SFT.

Labeling Experience

AI Developer Intern - Arch Technologies

Image

I supported the development of AI-powered solutions for customer-facing use cases during my internship at Arch Technologies. I worked on implementing AI chat assistants and automation components using Python and REST APIs. The role required strong hands-on skills in prompt engineering and integration of AI services into practical workflows. • Built AI chat assistant features for streamlined customer support. • Integrated automation tools and AI responses via Python and REST APIs. • Applied prompt engineering techniques to improve response quality and usability. • Collaborated on testing and refinement of AI components within delivery timelines.

2025 - 2026

Internee, AI development with medical LLM (QLoRA/PEFT) fine-tuning project work

Fine-tuning

Developed and managed a QLoRA fine-tuning workflow for a medical LLM based on Llama 3, including adapter and training lifecycle configuration. Performed domain-specific training setup in a notebook environment using 4-bit quantization for efficiency and resource reduction. Validated outputs as part of an end-to-end fine-tuning process suitable for downstream medical Q&A use. • Implemented PEFT lifecycle steps such as tokenization and adapter configuration • Used Unsloth 4-bit quantization to reduce VRAM overhead • Ran epoch-based training in Google Colab • Produced a specialized medical instruction-tuned model via fine-tuning

2025 - 2026

Education

Work History