Research Engineer Applied ML AI

@ Sully.ai

About Sully.ai

Our team comes from OpenAI, DeepMind, NASA, GoogleX, Tesla, and includes 2 physicians. With 6 exits, 2 IPOs, our model outperforms Claude, Gemini, and GPT-4.5 on clinical benchmarks and we have partnered with 400+ healthcare organizations in 16 months. We have raised $25M from YC, Amity Ventures, Sequoia scouts, among others and are targeting a $1T+ market opportunity.

About the Role

The Research Engineer Applied ML AI will bridge cutting-edge research and scalable production systems by owning training, fine-tuning, and inference toolchains. The role involves driving multimodal (text, audio, vision) integration and optimizing model throughput and evaluation in production. You will ensure Sully.ai's research artifacts translate into stable, high-performance features powering our healthcare agents.

Key Responsibilities

• Own full training, fine-tuning, and inference toolchains.
• Translate research repositories into production-ready services with stable APIs.
• Ship multimodal features (text, audio, vision) to boost agent performance.
• Optimize inference pipelines for cost, throughput, and latency.
• Build evaluation systems integrated into CI/CD to block weak checkpoints.

Hard Requirements

• Strong engineering background with experience in distributed systems and large-scale model training/serving.
• Hands-on experience with multimodal ML (audio, vision, text).
• Experience with production ML hygiene: versioning, metrics, observability, reproducibility.
• Proven track record of shipping ML systems into production.

Nice-to-Have

• Experience with model optimization techniques such as quantization, caching, pruning.
• Background in healthcare, medical AI, or other high-stakes regulated environments.
• Contributions to open-source ML frameworks or libraries.

First-Month Focus

• Audit and streamline current training, fine-tune, and inference pipelines.
• Stand up evaluation frameworks that gate deployments in CI.
• Deliver first improvements in throughput, cost, or latency of deployed models.

Success OKRs (90 Days)

• Deploy at least one multimodal feature (speech or vision) to production agents.
• Reduce inference cost or latency by 30% via optimization strategies.
• Integrate evaluation guardrails into CI/CD to block underperforming model releases.

Culture Fit

• Persistent, driven problem solver.
• Willing to push back on leadership to defend quality/timelines.
• Thrives in high-ambiguity, fast-paced startup environments.

Why Join Sully.ai?

• Shape the Future of Healthcare by building impactful partnerships.
• Enjoy early-stage impact and a key role in company growth.
• Remote-first culture with a flexible, mission-driven team.
• Competitive compensation including salary, equity, and growth opportunities.
• Solve scalability challenges in a rapidly growing company.

Key Skills/Competency

• ML Engineering
• Production ML
• Multimodal Integration
• Distributed Systems
• API Development
• CI/CD
• Pipeline Optimization
• Inference Systems
• Model Training
• Healthcare AI