Software Engineer, Formation Bio

• A track record of shipping production systems in ambiguous, fast-moving environments
• You build with AI tools daily. Claude Code, Copilot, agentic workflows - whatever the best current tools are, you're already using them and have opinions about what works
• Strong fundamentals: you write clean, maintainable code, you understand distributed systems and cloud infrastructure (AWS, GCP, or Azure), and you can debug hard problems
• You're drawn to the problem space, not just the stack. You'd rather understand why a clinical operations team needs a tool than debate framework choices in isolation
• Comfortable with ambiguity. You've worked in 0-to-1 settings or early-stage environments where the playbook didn't exist. That energizes you
• Excellent communicator. You can explain technical trade-offs to non-technical stakeholders and align a room around a decision
• You want to operate at a higher tempo and a higher standard than you have before, and you want to be surrounded by people who hold themselves to the same bar

Desirable

• We're hiring Software Engineers to help implement what engineering looks like at an AI-native pharma company. This is not a traditional SWE role and it's not an ML research role - it's for someone who builds production software in a world where AI has fundamentally changed what one engineer can do
• At Formation, the lines between engineering, product management, and data science are converging. Engineers talk directly to users, PMs write code, and domain experts across the company build and deploy real tools. We need engineers who thrive in that environment: people who move fast, care about the problem they're solving (not just the code), and see AI tools as a permanent force multiplier rather than a novelty
• You'll work across the drug development lifecycle - building systems that accelerate how we find, evaluate, develop, and deliver new medicines. You'll collaborate with clinical, regulatory, BD, and operations teams, not from behind a ticket queue, but embedded in the work. Some of what you build will be net-new products. Some will be infrastructure that enables non-engineers to build safely alongside you. All of it matters
• If you're energized by opportunity, and motivated by the idea that your work directly helps get medicines to patients faster, this is the role
• Design, build, and ship production software that accelerates drug development - from internal tools to user-facing products to platform capabilities
• Use AI tools (LLMs, agentic coding systems, retrieval-augmented pipelines) as a core part of how you work, not as an experiment on the side
• Get close to the business. Understand the clinical, regulatory, or operational problem you're solving well enough to make good product decisions without waiting for a spec
• Build for composability. Clean APIs, structured schemas, machine-readable documentation - what you build needs to be discoverable and extensible by people and agents who aren't engineers
• Review and enable contributions from non-engineers. PMs, designers, and domain experts across the company are shipping code. You help ensure quality, security, and maintainability
• Build the guardrails that lift technical quality. Proactively create automated tools - security scanners, skill files for end users, agents that clean up and standardize non-engineer code - that raise the floor of quality across the entire organization without slowing anyone down
• Prototype rapidly. Many ideas should be live and testable within days, not weeks. Build the tool, get feedback, iterate
• Collaborate across engineering, data science, product, and domain teams - the boundaries between these functions are fluid here, and we expect you to operate across them