Achira

The Future of Drug Discovery Powered by AI and Physics

Achira is a biotechnology startup that emerged from stealth in early 2025 with a $33 million Seed round backed by Dimension, Amplify Partners, NVIDIA's NVentures, and Compound. The company is pioneering a new approach to drug discovery by building atomistic foundation simulation models. By combining geometric deep learning, quantum chemistry, physics, and statistical mechanics, Achira aims to overcome the limitations of traditional biomolecular simulation and pure machine learning models. Their goal is to create large, accurate synthetic datasets free of experimental artifacts, transforming drug discovery into a true inverse design problem.

A Culture of Rigor and Scientific Depth

Achira fosters a culture that rewards rigor, speed, and scientific depth with an ownership mindset. The team comprises machine learners, quantum chemists, engineers, and computer-aided drug discovery practitioners. They operate at the frontier scale of large models and datasets, emphasizing high-throughput evaluation on an ML-framework-native biomolecular simulation stack. Employees are expected to take end-to-end ownership of their work, from model conception to the design of prospective tools, in an environment that values rapid iteration and cross-disciplinary collaboration.

What You'll Actually Do

Work at Achira involves tackling complex, frontier-scale problems at the intersection of AI and physical sciences.

- Machine Learning Researchers focus on designing probabilistic generative models (such as diffusion models, normalizing flows, and flow matching) for molecular systems. They develop efficient samplers and learned proposal mechanisms that adapt to complex conformational landscapes.

- Software Engineers build and scale the distributed infrastructure required to train and deploy these massive simulation models, working closely with researchers to implement robust, reproducible software in Python (PyTorch/JAX).

- Computational Chemists partner with the AI and platform teams to shape objectives related to drug potency, selectivity, and developability, running prospective design studies.

The work is highly collaborative, requiring systems thinking and the ability to integrate models into end-to-end pipelines.

Compensation & Benefits

Achira offers highly competitive compensation for its specialized roles. Based on public job postings, a Machine Learning Research Scientist focused on Atomistic Simulation Models can expect a base salary range of $165K to $259K annually. While specific details on equity and comprehensive benefits packages are not publicly detailed, early-stage startups of this caliber typically offer significant equity stakes (options or RSUs) to early employees, alongside standard health and wellness benefits.

The Interview Process

While specific interview rounds are not publicly documented, candidates can expect a rigorous process tailored to the highly technical nature of the work. Given the focus on probabilistic ML, statistical mechanics, and software engineering hygiene, the process likely involves deep technical screens assessing proficiency in PyTorch/JAX, understanding of generative modeling on structured data, and the ability to design efficient sampling algorithms. Candidates should be prepared to discuss their research track record, open-source contributions, and practical experience with large-scale ML systems.

Why Join / Why Not

Why Join: Achira offers a rare opportunity to work at the cutting edge of AI and drug discovery, backed by top-tier investors including NVIDIA. If you are passionate about applying probabilistic machine learning and statistical mechanics to solve fundamental problems in biomolecular design, this is an environment where you can have massive impact. The hybrid work model across San Francisco and New York provides flexibility and access to major tech hubs.

Why Not: The work is highly specialized and requires deep expertise in both machine learning and physical sciences; it is not a typical SaaS engineering environment. As an early-stage startup, the pace will be intense, and the path to clinical validation of AI-designed drugs is inherently long and risky. Those looking for highly structured, mature engineering processes or who prefer working on consumer-facing products might find the environment challenging.