Top Tech Jobs & Startup Jobs in NYC, NY

Serve as a technical product engineer advising customers on integrating Claude and LLMs into production products. Lead discovery, prototypes, eval frameworks, workshops, code reviews, and deployments while contributing patterns and documentation to product and engineering teams. Travel occasionally for customer support and represent Anthropic at events.

Lead and coordinate RL research programs: monitor experiment health and performance, prioritize and track experiments, run research reviews, establish lightweight processes, and resolve technical blockers across research, infrastructure, and data teams to accelerate model development.

As a Technical Cyber Threat Investigator, you will detect and counter misuse of AI systems, develop strategies against cyber threats, and collaborate across teams to improve safety measures.

Build and own claude.ai's serving and agent execution platform: sandboxed compute for safe code execution, state management for long-running agent tasks, delegated auth, observability for agent actions, and partner with product and research teams to design and ship scalable, secure infrastructure.

The Performance Engineer will analyze system performance, evaluate model outputs, build observability tools, and partner with teams to optimize throughput, latency, and correctness in AI systems.

Own end-to-end enablement for the Claude Developer Platform: build demos, hands-on labs, and self-service resources; deliver and scale live technical training; coach technical sellers; stress-test features with product; and measure content effectiveness to improve field readiness.

As an Engineering Manager, you'll lead efforts to enhance model performance, manage team projects, and support professional growth in a dynamic environment.

Build, maintain, and improve algorithms and infrastructure for finetuning and RL training (RLHF) of large language models. Improve speed, reliability, and usability of training pipelines through profiling, instrumentation, debugging GIL contention, launching test jobs, and implementing research algorithms to support production and internal model training.

Lead system-level architecture for large-scale AI compute: define specs, evaluate vendor designs, drive interconnect, power, thermal, and reliability strategies, develop validation and bring-up plans, and partner with ML, infrastructure, and supply chain teams.

Design and build real-time risk decisioning and tooling to prevent payments fraud. Implement fraud signals at scale, manage dispute and chargeback workflows, monitor metrics (loss, dispute, approval, false positives), lead investigations into emerging fraud patterns, and collaborate cross-functionally with finance, support, legal, data science, and external payment partners.