Python Expert for LLM Performance Improvement
Budget: ₹750 – ₹1,250 INR
Job Title: Top Python Coder
Location: India, Pakistan, Nigeria, Kenya, Egypt, Ghana, Bangladesh, Turkey, Mexico
Employment Type: Contract (3+ months, no medical/paid leave)
Working Hours: 8 hours per day / 40 hours per week (minimum 4 hours overlap with PST)
Start Date: Expected next week
Role Overview
This project is in collaboration with one of the foundational Large Language Model (LLM) companies. The goal is to improve LLM performance by generating, evaluating, and refining high-quality datasets.
You will work on creating, analyzing, and refining Python-based datasets that will be used for Supervised Fine-Tuning (SFT) and Reinforcement Learning with Human Feedback (RLHF).
This role does not involve training or building LLMs, but rather focuses on producing data that enhances their learning and performance.
Key Responsibilities
• Write efficient, high-quality Python code for AI model training and optimization tasks.
• Conduct model evaluations (Evals) to benchmark and analyze LLM performance.
• Evaluate and rank AI-generated responses based on technical accuracy, clarity, and alignment with task criteria.
• Create and explain reasoning-based evaluations and code-level improvements.
• Design and maintain high-quality datasets for Supervised Fine-Tuning (SFT) tasks.
• Collaborate with researchers and annotators on Reinforcement Learning with Human Feedback (RLHF) workflows.
• Participate in peer reviews of code and provide feedback for quality enhancement.
• Continuously research new tools and techniques to improve model evaluation and fine-tuning processes.
Requirements
• Minimum 3 years of software development experience, with a strong focus on Python.
• Excellent problem-solving and algorithmic skills (proven through LeetCode, HackerRank, GitHub, etc.).
• Strong ability to break down complex problems into logical, stepwise solutions.
• Clear and concise written communication in English.
• High attention to detail and ability to design thoughtful coding or reasoning tasks.
• Familiarity with AI/ML workflows or evaluation is a plus (not mandatory).
Location: India, Pakistan, Nigeria, Kenya, Egypt, Ghana, Bangladesh, Turkey, Mexico
Employment Type: Contract (3+ months, no medical/paid leave)
Working Hours: 8 hours per day / 40 hours per week (minimum 4 hours overlap with PST)
Start Date: Expected next week
Role Overview
This project is in collaboration with one of the foundational Large Language Model (LLM) companies. The goal is to improve LLM performance by generating, evaluating, and refining high-quality datasets.
You will work on creating, analyzing, and refining Python-based datasets that will be used for Supervised Fine-Tuning (SFT) and Reinforcement Learning with Human Feedback (RLHF).
This role does not involve training or building LLMs, but rather focuses on producing data that enhances their learning and performance.
Key Responsibilities
• Write efficient, high-quality Python code for AI model training and optimization tasks.
• Conduct model evaluations (Evals) to benchmark and analyze LLM performance.
• Evaluate and rank AI-generated responses based on technical accuracy, clarity, and alignment with task criteria.
• Create and explain reasoning-based evaluations and code-level improvements.
• Design and maintain high-quality datasets for Supervised Fine-Tuning (SFT) tasks.
• Collaborate with researchers and annotators on Reinforcement Learning with Human Feedback (RLHF) workflows.
• Participate in peer reviews of code and provide feedback for quality enhancement.
• Continuously research new tools and techniques to improve model evaluation and fine-tuning processes.
Requirements
• Minimum 3 years of software development experience, with a strong focus on Python.
• Excellent problem-solving and algorithmic skills (proven through LeetCode, HackerRank, GitHub, etc.).
• Strong ability to break down complex problems into logical, stepwise solutions.
• Clear and concise written communication in English.
• High attention to detail and ability to design thoughtful coding or reasoning tasks.
• Familiarity with AI/ML workflows or evaluation is a plus (not mandatory).