Software Engineer, Agent SDK - Claude Code
New York City, NY; San Francisco, CA | New York City, NY; Seattle, WAFull-TimeMid-levelSoftware Engineering
About Anthropic
- Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.
About Anthropic
- Anthropic's mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.
The Role
- The Claude Agent SDK is the foundational layer that powers autonomous AI agents — giving developers access to the same tools, agent loop, and context management that power Claude Code, programmable in Python and TypeScript. It's what enables agents to read files, run commands, search the web, edit code, and more, out of the box.
- We're looking for a Software Engineer to own and lead the Agent SDK. This is a high-ownership, high-autonomy role: you'll set the technical direction for the SDK, be opinionated about its API design and developer experience, and be the connective tissue between internal teams building next-generation experiences (Claude Desktop, Cowork, IDE extensions) and external developers building production agents on top of Claude.
- You'll need to exercise taste and judgment about what to build — balancing a rich stream of requests from internal stakeholders and external customers against a clear product vision for what a great agentic SDK should be.
Responsibilities
- Own the roadmap and technical direction for the Claude Agent SDK (Python and TypeScript)
- Design clean, intuitive APIs that make it easy to build powerful, reliable agents
- Partner closely with internal teams building on Agent SDK — including Claude Desktop, Cowork, and IDE extensions — to understand their needs and prioritize accordingly
- Engage directly with external developers to understand how they're using the SDK, where they're hitting friction, and what unlocks the next generation of use cases
- Raise the quality bar on documentation, examples, and the overall developer experience
- Collaborate with other Claude Code engineers to expose new model capabilities through clean SDK abstractions
You may be a good fit if you
- Have strong opinions about what great developer tools look like, grounded in experience building or using them
- Are energized by owning something end-to-end, from API design through docs through developer adoption
- Have experience designing and shipping developer-facing SDKs or APIs, and have felt the difference between an SDK that developers love and one they merely tolerate
- Are excited about the frontier of agentic AI — agents that can reason, plan, and act autonomously
- Have hands-on experience with agentic patterns: tool use, orchestration, context management, multi-step reasoning
- Can operate independently in ambiguous environments and push projects forward without waiting for direction
- Care about the societal implications of the technology you build
- Have 5+ years of software engineering experience
Strong candidates may also have
- Maintained a project built on the Claude Agent SDK
- Experience managing an open source project on GitHub — triaging issues, prioritizing what to take on and engaging with community contributors
- Experience scaling TypeScript or Python library codebases
- Shipped SDKs or APIs used by external developers in production
- Opinions about API versioning, backwards compatibility, and managing ecosystem migrations
- The annual compensation range for this role is listed below.
- For sales roles, the range provided is the role’s On Target Earnings ("OTE") range, meaning that the range includes both the sales commissions/sales bonuses target and annual base salary for the role.
How we're different
- We believe that the highest-impact AI research will be big science. At Anthropic we work as a single cohesive team on just a few large-scale research efforts. And we value impact — advancing our long-term goals of steerable, trustworthy AI — rather than work on smaller and more specific puzzles. We view AI research as an empirical science, which has as much in common with physics and biology as with traditional efforts in computer science. We're an extremely collaborative group, and we host frequent research discussions to ensure that we are pursuing the highest-impact work at any given time. As such, we greatly value communication skills.
- The easiest way to understand our research directions is to read our recent research. This research continues many of the directions our team worked on prior to Anthropic, including: GPT-3, Circuit-Based Interpretability, Multimodal Neurons, Scaling Laws, AI & Compute, Concrete Problems in AI Safety, and Learning from Human Preferences.
