Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I'm an AI/ML engineer currently pursuing an MSc in Data Science & Data Analytics at University College Cork — two semesters in, holding a 1:1 with a 73% aggregate — after finishing a B.Tech in Artificial Intelligence and Data Science with a First Class GPA of 8.47. Over the last two years I've worked across the AI stack — fine-tuning small language models with LoRA/PEFT, deploying quantized GGUF models, building production RAG pipelines with LangGraph, and shipping multi-agent systems that automate real engineering work. What drives me is using AI to solve hard, real problems — building and shipping deep learning models across domains, pushing the limits of what runs on consumer hardware, and turning messy research into working systems. Agentic and multi-agent systems are part of my toolkit, not the ceiling.…I'm an AI/ML engineer currently pursuing an MSc in Data Science & Data Analytics at University College Cork — two semesters in, holding a 1:1 with a 73% aggregate — after finishing a B.Tech in Artificial Intelligence and Data Science with a First Class GPA of 8.47. Over the last two years I've worked across the AI stack — fine-tuning small language models with LoRA/PEFT, deploying quantized GGUF models, building production RAG pipelines with LangGraph, and shipping multi-agent systems that automate real engineering work. What drives me is using AI to solve hard, real problems — building and shipping deep learning models across domains, pushing the limits of what runs on consumer hardware, and turning messy research into working systems. Agentic and multi-agent systems are part of my toolkit, not the ceiling.

Ashfaq Ahamed

AI Engineer, Data Scientist, Software Engineer, +2





I'm an AI/ML engineer currently pursuing an MSc in Data Science & Data Analytics at University College Cork — two semesters in, holding a 1:1 with a 73% aggregate — after finishing a B.Tech in Artificial Intelligence and Data Science with a First Class GPA of 8.47. Over the last two years I've worked across the AI stack — fine-tuning small language models with LoRA/PEFT, deploying quantized GGUF models, building production RAG pipelines with LangGraph, and shipping multi-agent systems that automate real engineering work. What drives me is using AI to solve hard, real problems — building and shipping deep learning models across domains, pushing the limits of what runs on consumer hardware, and turning messy research into working systems. Agentic and multi-agent systems are part of my toolkit, not the ceiling.…I'm an AI/ML engineer currently pursuing an MSc in Data Science & Data Analytics at University College Cork — two semesters in, holding a 1:1 with a 73% aggregate — after finishing a B.Tech in Artificial Intelligence and Data Science with a First Class GPA of 8.47. Over the last two years I've worked across the AI stack — fine-tuning small language models with LoRA/PEFT, deploying quantized GGUF models, building production RAG pipelines with LangGraph, and shipping multi-agent systems that automate real engineering work. What drives me is using AI to solve hard, real problems — building and shipping deep learning models across domains, pushing the limits of what runs on consumer hardware, and turning messy research into working systems. Agentic and multi-agent systems are part of my toolkit, not the ceiling.

Available to hire

I’m an AI/ML engineer currently pursuing an MSc in Data Science & Data Analytics at University College Cork — two semesters in, holding a 1:1 with a 73% aggregate — after finishing a B.Tech in Artificial Intelligence and Data Science with a First Class GPA of 8.47.

Over the last two years I’ve worked across the AI stack — fine-tuning small language models with LoRA/PEFT, deploying quantized GGUF models, building production RAG pipelines with LangGraph, and shipping multi-agent systems that automate real engineering work.

What drives me is using AI to solve hard, real problems — building and shipping deep learning models across domains, pushing the limits of what runs on consumer hardware, and turning messy research into working systems. Agentic and multi-agent systems are part of my toolkit, not the ceiling.

Skills

Experience Level

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Expert

Intermediate

Language

Afar

Advanced

Amharic

Intermediate

Work Experience

AI Engineer Intern at Tickingminds

January 1, 2025 - August 1, 2025

Fine-tuned privacy-focused Small Language Models using LoRa/PEFT and Unsloth with custom-curated datasets; optimized and deployed quantized GGUF models using llama.cpp, reducing reliance on external closed-source LLM APIs by ~70% and improving enterprise privacy compliance. Designed and deployed a production-grade RAG pipeline using LangGraph for retrieval and re-ranking over historical company archives, improving contextual access accuracy by ~40%, and deployed on AWS EC2 as part of the company’s TSigma AI platform. Engineered a multi-agent Claude-based automated testing system integrated with a custom Playwright MCP server (WebSockets + JSON-RPC) to automate end-to-end test execution, ATS generation, and storage to Amazon S3—reducing manual QA effort by ~60%.

Machine Learning Intern at Phooldaan

June 1, 2024 - September 1, 2024

Developed a hybrid image-based retrieval system combining Generative AI and Deep Learning to improve product discovery and visual search relevance on the e-commerce platform. Built a multimodal search pipeline using Gemini API to generate contextual image descriptions (texture, color, decor style, setting) and pretrained EfficientNet CNNs to extract visual features; stored semantic and visual embeddings in a vector database for similarity-based retrieval. Engineered end-to-end image search workflow using both AI-generated metadata and CNN embeddings to retrieve visually and contextually similar products, improving retrieval accuracy by ~45% and reducing irrelevant results.

Data Science Intern at Cognifyz Technologies

January 1, 2024 - February 1, 2024

Performed data exploration and preprocessing on a restaurant dataset (10k+ records), including missing value handling and data type conversions. Analyzed the distribution of the target variable (Aggregate rating) to assess class imbalance and conducted descriptive analysis across numeric features, improving analytical reporting efficiency by ~25%. Explored distributions of categorical variables to identify top cuisines and cities with the highest number of restaurants, producing useful insights into regional dining trends.