Software Engineer, Memory & Search
New York City, NY; New York City, NY | Seattle, WA; Remote-Friendly, United States; San Francisco, CA | New York City, NY | Seattle, WA·Posted today
aillmtypescriptpython
<div class="content-intro"><h2><strong>About Anthropic</strong></h2> <p>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.</p></div><h2 class="text-text-100 mt-3 -mb-1 text-[1.125rem] font-bold" data-sourcepos="5:1-5:18;181-198">About the role</h2> <p class="font-claude-response-body break-words whitespace-normal" data-sourcepos="7:1-7:475;200-674">Anthropic is looking for Software Engineers to join the Memory & Search team. We own how Claude remembers you and how it finds what it needs, across Claude.ai, Cowork and Claude Code. That spans the user-facing experience (what Claude remembers, how people see and edit it, how search surfaces the right conversation) and the shared infrastructure underneath it, the memory service, indexing, retrieval and vector search.</p> <p class="font-claude-response-body break-words whitespace-normal" data-sourcepos="7:1-7:475;200-674">Memory is one of the biggest levers on whether Claude feels like it knows you. We recently moved memory from a periodically synthesized summary to a dynamic, editable, on-demand system, and then unified it so the same memory follows you across products. Search went from a top user complaint to something people rely on to find past work. The platform is now in place, and the next phase is finding the product form factors that make memory the most compelling and sticky part of using Claude.</p> <p class="font-claude-response-body break-words whitespace-normal" data-sourcepos="7:1-7:475;200-674">That is why we are hiring product engineers. The team is strong on infrastructure, and we want people who have shipped user-facing products at scale, have a sharp sense for what makes a product feel good, and have built agentic systems where the model, the tools and the user interface have to work together. You will run short, demo-driven experiments on top of the memory and search stack, decide what to keep based on real usage, and help the whole team think like a product team. If you have founded a company, been an early engineer at a startup, or taken a product from zero to one, you will recognize the way we want to work.</p> <p class="font-claude-response-body break-words whitespace-normal" data-sourcepos="7:1-7:475;200-674">This is a trust-sensitive surface. Memory touches privacy, consent and enterprise controls, so you will work closely with product, design, privacy, legal and safeguards, and you will be expected to treat those constraints as part of good product design rather than obstacles to it.</p> <h2 class="text-text-100 mt-3 -mb-1 text-[1.125rem] font-bold" data-sourcepos="11:1-11:24;1119-1142">Key responsibilities</h2> <ul class="[li_&]:mb-0 [li_&]:mt-1 [li_&]:gap-1 [&:not(:last-child)_ul]:pb-1 [&:not(:last-child)_ol]:pb-1 list-disc flex flex-col gap-1 pl-8 mb-3" data-sourcepos="13:1-17:88;1144-1746"> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="13:1-13:138;1144-1281">Design, build and ship memory and search experiences in Claude.ai, Cowork and Claude Code, owning them end to end from the backend service through the model's tools to the user interface</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="13:1-13:138;1144-1281">Run product experiments on the memory and search stack: form a hypothesis, build a demo, put it in front of real users, measure, and make a clear persevere-or-pivot call</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="13:1-13:138;1144-1281">Shape how Claude uses memory and retrieval inside its agentic loop, including what the model stores, when it searches, how results are ranked and presented, and how the user stays in control</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="13:1-13:138;1144-1281">Define and track the metrics that tell us whether memory is actually making Claude more useful, and build the evaluations that keep quality honest as the system evolves</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="13:1-13:138;1144-1281">Improve search relevance and recall across conversations and memories, working across embeddings, hybrid retrieval and ranking as well as the interface that presents results</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="13:1-13:138;1144-1281">Partner with product, design, privacy, legal and safeguards to ship features that respect user consent and enterprise requirements such as customer-managed encryption keys and data retention policies</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="13:1-13:138;1144-1281">Raise the bar on product quality for the team through design reviews, code reviews and the standards you set for latency, polish and reliability<br>Share the on-call for the systems you own</li> </ul> <h2 class="text-text-100 mt-3 -mb-1 text-[1.125rem] font-bold" data-sourcepos="19:1-19:26;1748-1773">Minimum qualifications</h2> <ul class="[li_&]:mb-0 [li_&]:mt-1 [li_&]:gap-1 [&:not(:last-child)_ul]:pb-1 [&:not(:last-child)_ol]:pb-1 list-disc flex flex-col gap-1 pl-8 mb-3" data-sourcepos="21:1-25:72;1775-2196"> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="21:1-21:89;1775-1863">Experience building and operating user-facing products used by a large number of people, including the backend services behind them</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="21:1-21:89;1775-1863">Experience building agentic systems or LLM-powered products, including tool use, retrieval and prompt or context design</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="21:1-21:89;1775-1863">Strong programming skills in Python and TypeScript, with experience across backend services and web front ends</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="21:1-21:89;1775-1863">A track record of taking a product or feature from an idea to launch, and iterating on it based on usage data and user feedback</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="21:1-21:89;1775-1863">Experience defining product metrics and running experiments to make decisions</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="21:1-21:89;1775-1863">Strong product judgment: you notice when something feels off and you fix it, and you can explain why a design choice matters to users</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="21:1-21:89;1775-1863">Strong written and verbal communication, with experience working closely with product managers, designers and non-engineering partners</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="21:1-21:89;1775-1863">Interest in AI safety and Anthropic's mission</li> </ul> <h2 class="text-text-100 mt-3 -mb-1 text-[1.125rem] font-bold" data-sourcepos="27:1-27:28;2198-2225">Preferred qualifications</h2> <ul class="[li_&]:mb-0 [li_&]:mt-1 [li_&]:gap-1 [&:not(:last-child)_ul]:pb-1 [&:not(:last-child)_ol]:pb-1 list-disc flex flex-col gap-1 pl-8 mb-3" data-sourcepos="29:1-35:75;2227-2767"> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="29:1-29:83;2227-2309">Experience as a founder, early-stage engineer, or someone who has taken a product from zero to one</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="29:1-29:83;2227-2309">Experience building search or retrieval systems: embeddings, vector search, hybrid lexical and semantic retrieval, or ranking</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="29:1-29:83;2227-2309">Experience building evaluations for LLM or ML systems, or measuring the quality of retrieval and ranking</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="29:1-29:83;2227-2309">Experience on consumer products with strong retention, or on personalization systems</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="29:1-29:83;2227-2309">Experience shipping features that handle sensitive data or meet enterprise requirements such as customer-managed encryption keys, data residency or HIPAA</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="29:1-29:83;2227-2309">Experience with latency-sensitive, performance-critical product work</li> <li class="font-claude-response-body whitespace-normal break-words pl-2" data-sourcepos="29:1-29:83;2227-2309">Experience mentoring engineers or setting technical direction for a team</li> </ul><div class="content-pay-transparency"><div class="pay-input"><div class="description"><p>The annual compensation range for this role is listed below. </p> <p>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.</p></div><div class="title">Annual Salary:</div><div class="pay-range"><span>$320,000</span><span class="divider">—</span><span>$485,000 USD</span></div></div></div><div class="content-conclusion"><h2><strong>Logistics</strong></h2> <p><strong>Minimum education: </strong>Bachelor’s degree or an equivalent combination of education, training, and/or experience</p> <p><strong>Required field of study: </strong>A field relevant to the role as demonstrated through coursework, training, or professional experience</p> <p><strong>Minimum years of experience: </strong>Years of experience required will correlate with the internal job level requirements for the position</p> <p><strong>Location-based hybrid policy:</strong> Currently, we expect all staff to be in one of our offices at least 25% of the time. However, some roles may require more time in our offices.</p> <p><strong data-stringify-type="bold">Visa sponsorship:</strong> We do sponsor visas! However, we aren't able to successfully sponsor visas for every role and every candidate. But if we make you an offer, we will make every reasonable effort to get you a visa, and we retain an immigration lawyer to help with this.</p> <p><strong>We encourage you to apply even if you do not believe you meet every single qualification.</strong> Not all strong candidates will meet every single qualification as listed. Research shows that people who identify as being from underrepresented groups are more prone to experiencing imposter syndrome and doubting the strength of their candidacy, so we urge you not to exclude yourself prematurely and to submit an application if you're interested in this work. We think AI systems like the ones we're building have enormous social and ethical implications. We think this makes representation even more important, and we strive to include a range of diverse perspectives on our team.<br><br><strong data-stringify-type="bold">Your safety matters to us.</strong> To protect yourself from potential scams, remember that Anthropic recruiters only contact you from @anthropic.com email addresses. In some cases, we may partner with vetted recruiting agencies who will identify themselves as working on behalf of Anthropic. Be cautious of emails from other domains. Legitimate Anthropic recruiters will never ask for money, fees, or banking information before your first day. If you're ever unsure about a communication, don't click any links—visit <u data-stringify-type="underline"><a class="c-link c-link--underline" href="http://anthropic.com/careers" target="_blank" data-stringify-link="http://anthropic.com/careers" data-sk="tooltip_parent" data-remove-tab-index="true">anthropic.com/careers</a></u> directly for confirmed position openings.</p> <h2><strong>How we're different</strong></h2> <p>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.</p> <p>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.</p> <h2><strong>Come work with us!</strong></h2> <p>Anthropic is a public benefit corporation headquartered in San Francisco. We offer competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and a lovely office space in which to collaborate with colleagues. <strong data-stringify-type="bold">Guidance on Candidates' AI Usage:</strong> Learn about <a class="c-link" href="https://www.anthropic.com/candidate-ai-guidance" target="_blank" data-stringify-link="https://www.anthropic.com/candidate-ai-guidance" data-sk="tooltip_parent">our policy</a> for using AI in our application process.</p></div>