Artificial Intelligence Developer Intern

PiRails Private Limited; Lahore; On-site; PKR 508,000 - 608,000; Full time

Job summary

PiRails Private Limited is seeking a motivated Artificial Intelligence Developer Intern to help build, optimize, and deploy AI solutions across local and edge environments in Lahore. The role involves hands-on work with open-source models and lightweight deployments, with a focus on practical impact and learning in a full-time internship.

Responsibilities include training pipelines, model evaluation, and collaboration with backend and hardware engineers to integrate AI features into end-user

Qualifications

• Strong foundation in machine learning, deep learning, and practical ML workflows.
• Experience with open-source models and evaluation on real-world data.
• Familiarity with edge deployment frameworks and lightweight models.

Responsibilities

• Design, train, and fine-tune AI models using open-source datasets and frameworks.
• Experiment with open-weight models (e.g., Llama, Mistral, Whisper) to solve real tasks.
• Optimize and deploy lightweight models on edge devices or client platforms.
• Collaborate with backend/hardware teams to integrate AI solutions end-to-end.

Skills

Python

Machine learning

Deep learning

Edge AI

Open-source

Model evaluation

Tools

PyTorch

TensorFlow

JAX

Hugging Face

ONNX Runtime

Transformers.js

Job description

Artificial Intelligence Developer Intern

Artificial Intelligence Developer Intern

place Township, Lahore

work Full-time

payments 50,000

Job Description

Role Overview

We are seeking a driven Artificial Intelligence Developer intern to assist in building, optimizing, and deploying AI solutions across local and edge environments. In this role, you will work directly with open-source and open-weight models, custom-train architecture for specific tasks, and tackle the challenges of deploying lightweight AI models on resource-constrained edge devices.

Key Responsibilities

Model Training & Fine-Tuning: Design, build, and train machine learning models. Execute custom training and fine-tuning pipelines using modern open-source datasets and frameworks.

Open-Source & Open-Weight Ecosystem: Work extensively with state-of-the-art open-source and open-weight models (e.g., Llama, Mistral, Whisper, Flux, or Stable Diffusion) to solve real-world problems.

Edge Deployment: Quantize, optimize, and deploy lightweight models onto client-side and edge platforms using frameworks like Transformers.js, ONNX Runtime, TensorRT, or WebGPU.

Model Optimization: Apply optimization techniques such as pruning, quantization (INT8/INT4), distillation, and compilation to minimize memory footprint and latency while maintaining accuracy.

Cross-Functional Collaboration: Partner with backend and hardware engineers to integrate embedded/edge models into end-user applications and hardware pipelines.

Requirements & Qualifications

Core AI/ML Skills: Strong foundation in machine learning, deep learning principles, and frameworks such as PyTorch, TensorFlow, or JAX.

Open-Source Familiarity: Practical experience exploring, fine-tuning, and evaluating models from Hugging Face and other open-source repositories.

Edge & Web AI: Familiarity with edge runtime environments (e.g., Transformers.js, ONNX Runtime Web/Mobile, TFLite).

We typically respond to applicants within 2-3 business days.