Machine Learning Infrastructure Engineer
Aikium Inc. · Berkeley, CA · Full-Time
Build the Backbone of AI-Powered Protein Design
Aikium Inc. is transforming protein engineering with AI at scale. Our Yotta Display™ platform enables the synthesis and screening of trillion-scale protein libraries—redefining how therapeutics are discovered. We focus on designing several protein scaffolds (including novel non-antibody SeqRs) that bind intrinsically disordered regions of target proteins, with applications in oncology, inflammation, and beyond.
Behind every model, every scaffold, every breakthrough—there is infrastructure. And that’s where you come in.
Your Mission
We’re hiring a generalist Machine Learning Infrastructure Engineer to architect the foundation for Aikium’s data and inference systems. You will own the pipelines and backend tools that allow ML scientists and wet lab biologists to collaborate efficiently—by building robust data workflows, scalable inference pipelines, and accessible internal APIs.
This is a greenfield role. You will define how data flows from our high-throughput wet lab experiments to downstream model inference and reporting—cleanly, reliably, and at scale.
What You'll Do
Data Stewardship
- Design and implement pipelines to ingest, clean, and serve high-volume Yotta display data (~100M NGS reads / week)
- Integrate internal and public experimental datasets and model-generated data into unified storage and query-able formats
- Own schema design, version control, and metadata tracking for internal and external datasets
- Migrate the team off fragmented CSV workflows to a structured data architecture
Inference Infrastructure
- Build and maintain scalable cloud-based inference systems for batch and real-time model outputs
- Optimize inference performance using tools such as ONNX, TensorRT, or quantization libraries
- Deploy APIs that expose models to downstream applications or user-facing tools
Enablement & Internal Tooling
- Develop APIs (FastAPI preferred) that enable ML engineers and wet lab scientists to access data and inference services
- Build internal dashboards and lightweight UIs for data inspection, monitoring, and debugging
- Maintain clean documentation and onboarding guides for team-wide self-service
Reliability & Monitoring
- Implement logging, observability (e.g., Prometheus, Grafana), and robust failure handling
- Ensure data reproducibility and system stability as biological throughput scales
You Might Be a Fit If You...
- Have 3+ years of experience in ML infrastructure, backend engineering, or data platform roles
- Have built and maintained production data pipelines or inference APIs in Python or similar stacks
- Are comfortable navigating AWS (S3, ECS/Fargate, Lambda, etc.) and selecting the right tools for the job
- Thrive in startup environments where autonomy, speed, and clarity matter more than process
- Enjoy helping others succeed by making systems usable, discoverable, and debuggable
- Are proactive, collaborative, and service-oriented—especially when working with scientists outside your domain
Bonus Points
- Familiarity with biological data formats or lab informatics systems
- Experience working with ML inference tooling (ONNX, TorchScript, Triton Inference Server)
- Contributions to open source infra or bio tooling projects
Why Aikium?
- Be the first dedicated infrastructure hire at a fast-moving biotech AI company
- Architect greenfield systems with lasting impact on data and discovery velocity
- Work on a cutting-edge platform generating and interpreting trillions of protein sequences
- Competitive salary and meaningful equity; benefits include health, vision & dental insurance and 401(k)
---
Aikium Inc. is based in Berkeley, CA. We are an equal opportunity employer and are committed to building a diverse and inclusive team.
If you're excited to build the data and inference backbone of a new kind of AI-driven biotech company, we want to hear from you.