Role: Senior GenAI Engineer

Function: Engineering

Location: Remote

Type: Full-time

Compensation: Competitive salary starting at 50 LPA with unlimited upside potential based on experience, plus significant equity participation (ESOPs)

Industry: HR Technology, AI/ML, SaaS

About Company

A global HR platform that simplifies international hiring and workforce management for companies building distributed teams. The company serves as an Employer of Record (EOR) across 140+ countries.

It streamlines compliance, payroll, and benefits through a centralized platform. Founded by IIT Bombay and ISB alumni with deep AI/ML and operations expertise.

The company operates in a fast-paced startup environment that values transparency, compliance, and cross-border collaboration.

Position Overview

You will lead the fine-tuning of open-source generative AI models that power HR automation for a global workforce platform. You'll experiment with cutting-edge LLMs to create specialized solutions for compliance and workforce management processes. You'll have access to high-performance GPU infrastructure and the opportunity to shape AI-driven HR solutions used across 140+ countries.

Role & Responsibilities

• Fine-tune open-source LLMs including Llama, Mistral, and Falcon
• Experiment with different fine-tuning techniques including LoRA, QLoRA, and full parameter tuning
• Design and implement custom training pipelines using PyTorch for model optimization
• Evaluate model performance and conduct A/B testing on fine-tuned models
• Optimize model inference performance and memory usage for production deployment
• Research and implement state-of-the-art fine-tuning methodologies and techniques
• Collaborate with data teams to prepare and curate training datasets for model fine-tuning

Must Have Criteria

• 4+ years of hands-on experience with PyTorch for deep learning model development
• Proven hands-on experience fine-tuning Small Language Models (SLMs) using any technique (full fine-tuning, LoRA, QLoRA, PEFT)
• Strong expertise in parameter-efficient fine-tuning methods and their practical implementation
• Experience with Hugging Face Transformers library and PEFT library for model fine-tuning
• Proficiency in Python and deep learning frameworks for model experimentation
• Understanding of transformer architectures and attention mechanisms
• Experience with GPU computing and model training optimization

Nice to Have

• Knowledge of quantization techniques and model compression methods
• Experience with MLflow, Weights & Biases, or similar experiment tracking tools
• Previous startup or research environment experience with model experimentation
• Publications or contributions to open-source fine-tuning projects

What We Offer

• Competitive salary starting at 50 LPA with unlimited upside potential based on experience
• Significant equity participation (ESOPs) with high growth potential
• Opportunity to experiment with cutting-edge open-source AI models
• Remote-friendly work culture with flexible hours and autonomy
• Access to high-performance GPU infrastructure for model training