We're are looking for a Senior/Staff Full Stack Engineer with 10+ YOE to join our Engineering team. We need a builder who is excited to create foundational systems and have meaningful influence on our technology as we work to accelerate drug discovery and improve human health globally. This is a top priority hire driven by our hypergrowth and the need to scale our core product to meet surging enterprise demand.
Responsibilities
Build new product features and systems from scratch across frontend and backend, directly expanding Mithril's platform capabilities for world-class researchers
Design, build, and own scalable backend services, APIs, and data pipelines that power the Mithril AI platform across the application, identity, data storage, and ML layers
Work within fast-moving sprint cycles to ship code quickly, contributing to production within your first few weeks
Collaborate with a cross-functional team of engineers, product managers, designers, and AI scientists to translate complex scientific workflows into intuitive, high-performance software
Own features end-to-end from early development through production deployment, iteration, and scale — ensuring they deliver real value to researchers at leading pharma and biotech organizations
Mithrl is building the scientific decision engine for R&D, which it describes as the world's first commercially available AI Co-Scientist. Its platform turns messy, unstructured biological data into defensible insights in minutes instead of weeks or months. That gives pharma and biotech researchers back the time they would otherwise spend waiting on analysis pipelines, so they can focus on discovery.
Mission: We imagine a world where new medicines reach patients in months, not years, and where scientific breakthroughs happen at the speed of thought. The mission has not changed since day one: be the discovery engine for scientific breakthroughs and speed up drug discovery to improve human health around the world.
The problem: In traditional drug discovery, researchers run experiments that produce terabytes of unstructured data, such as DNA and RNA sequencing. They then hand that data to separate bioinformatics teams and often wait 3 to 8 weeks for insights from a single experiment. This bottleneck is one reason a pharma giant can take 8 to 10 years to land one drug IP.
The solution: Using only natural language, scientists upload their data to Mithrl and get real analysis, novel targets, hypotheses, and patent-ready reports, usually within 3 to 15 minutes. A proprietary biological knowledge graph powers the platform. It links how DNA and RNA interact with proteins, diseases, and existing drugs, and combines customer data with globally published research. Every step is transparent and reproducible.
Traction:
- 12x year-over-year revenue growth: from $0 to $2M in its first year live, and on track to pass its $10M goal this year
- Trusted by leading pharma and biotech companies, including Johnson & Johnson, Roche, Eli Lilly, AstraZeneca, and Siemens, across North America, Europe, and Australia
- Most early enterprise customers came back on their own to expand their contracts by a large margin
- Tied to six new drug IPs in its first year
- Backed by a recently closed Series A led by Obvious Ventures, a fund focused on deep tech and biotech
Culture: Mithrl is a tight-knit, talent-dense, in-person team in San Francisco. It values curiosity, consistency, clarity, and focused daily execution. The company ships quickly, about twice a week, and keeps improving based on real feedback from users.
Team size
25 employeesFounded
Website
mithrl.com/Total funding
Company locations
Mithrl has about 25 people, all working in person at the San Francisco office near Montgomery station. Both co-founders are software engineers. They worked across startups and larger companies such as NVIDIA, then moved into biocomputational engineering through their master's programs. The CEO, Vivek, is closely involved in building the product and hiring.
Engineering team: About 10 engineers, including a newly hired Head of Engineering, four full stack engineers, two data scientists, and infrastructure and DevOps engineers. The team also includes a product leader with a life sciences background and product designers. One of the company's very first hires was a product designer. Engineering plans to grow to 30 by the end of this year and to 60–70 the year after, which opens real paths to tech lead and management roles.
What it's like to work here:
- Create your own adventure: There is a lot to build. Engineers can lean toward customer-facing product work, backend and infrastructure at scale, or ML and knowledge graph systems.
- Clear direction: The founders know exactly what needs to be built over the next two years, so the ambiguity is about execution speed, not guessing what will work.
- High-energy, in-person culture: There is plenty of laughter, frequent product discussions, and team offsites. The team chooses to be in the office five days a week, with flexibility for those who need it.
- Not a 996 culture: Several leaders are parents, and the team values sustainable, focused work over clocking hours.
- Curiosity is celebrated: People who ask endless questions about the product and enjoy learning a new field do well here. No biology background is needed.
- Ship fast: Releases go out about twice a week, shaped by real feedback from world-class researchers.
Mithrl is an AI-powered scientific decision engine, or AI Co-Scientist, for pharma and biotech R&D. Researchers drag and drop up to 10 data files, such as omics or sequencing data, and ask questions in plain language. Mithrl then builds custom analysis workflows on demand to align, analyze, interpret, and combine complex datasets, with no manual coding or repetitive setup. Results arrive in about 3 to 15 minutes instead of the weeks or months a traditional bioinformatics hand-off takes. A proprietary biological knowledge graph powers the platform. It maps how DNA and RNA across humans and other organisms interact with proteins, diseases, and existing drugs, and it layers a customer's proprietary data on top of globally published research. Mithrl takes a "white box" approach: scientists can trace every step the AI took, inspect the underlying data transformations, and see exactly how it reached its conclusions, which keeps results reproducible. Through a chat-style interface, researchers can follow up with questions like "find me novel targets for this disease" or "why did this batch fail?" and get analysis, hypotheses, novel targets and biomarkers, and patent-ready reports. Enterprise customers such as Johnson & Johnson, Roche, Eli Lilly, AstraZeneca, and Siemens buy it on a usage-based model, paying for bundles of uploads.