Member of Technical Staff – Post-Training

This is a fully remote position, open to applicants in United States, +2 more countries.

📋 Description

• Create systems and methodologies for post-training of frontier models, which encompass supervised fine-tuning, reinforcement learning, preference optimization, reward modeling, and similar techniques.

• Convert open-ended research questions or partner requirements into hypotheses, experimental designs, evaluation strategies, and production-ready implementations.

• Develop and enhance evaluation frameworks, benchmarks, training environments, data-processing pipelines, and quality-control systems.

• Execute rapid iteration cycles: prototype, assess, analyze results, and apply insights to the next system or product.

• Collaborate with AI researchers and domain specialists to generate high-quality data, feedback, and evaluation methodologies.

• Recognize repeatable patterns across projects and transform them into reusable software and platforms.

• Elevate the technical standards through design judgment, effective communication, code quality, and mentorship.

• Contribute to benchmarks, open-source projects, research endeavors, and technical documentation where it provides leverage.

• Assist in shaping Handshake Labs' technical vision, operational culture, and reusable systems.

⛳️ Requirements

• A minimum of 3 years of proven experience in post-training, fine-tuning, or model-evaluation tasks.

• Relevant expertise may involve RL, SFT, LoRA/PEFT, full fine-tuning, RLHF, DPO, PPO, reward modeling, or training environments.

• Proficient in Python with the capability to write clean, efficient, and scalable code.

• Practical experience with contemporary ML tooling, especially PyTorch and large-scale data, training, or evaluation workflows.

• Strong experimental judgment, encompassing hypothesis formulation, metric selection, failure diagnosis, and differentiating signal from noise.

• Experience in designing systems and making trade-offs concerning quality, scalability, reliability, and reusability.

• Ability to thrive in a fast-paced, ambiguous environment with significant ownership.

• Strong collaborative communication skills and the ability to engage with researchers, engineers, domain experts, and customers.

• Particularly noteworthy: experience with large-scale ML training, inference, data, or evaluation systems.

• Especially compelling: work with LLM/agent benchmarks, evaluation methodologies, annotation systems, or data-quality frameworks.

• Especially noteworthy: involvement in reinforcement learning, alignment, model behavior, synthetic data, or human-in-the-loop systems.

• Especially compelling: published research, significant open-source contributions, or technical leadership in ML systems or AI research.

• Particularly attractive: translating research or repeated client projects into robust, reusable platforms.

🏝️ Benefits

• Equity in a rapidly-growing company.

• 401(k) matching.

• Competitive salary package.

• Financial coaching services.

• Paid parental leave.

• Fertility assistance.

• Parental coaching resources.

• Comprehensive medical, dental, and vision insurance.

• Mental health support services.

• $500 wellness stipend.

• $2,000 learning stipend.

• Ongoing professional development opportunities.

• Commuting assistance.

• Complimentary lunch.

• On-site gym facilities in the San Francisco office.

• Flexible paid time off.

• 15 holidays plus 2 flexible days.

• Team outings.

• Referral bonuses.