Top AI Engineer Jobs

Build and adapt AI models for customer-specific scientific workflows. Lead post-training using SFT and reinforcement learning, design evaluation systems, analyze model failures, and drive iterative quality improvements. Partner with AI researchers and software engineers to integrate reliable model capabilities into production workflows. The role requires strong Python and PyTorch skills, deep familiarity with language, multimodal, or agentic models, and the ability to translate customer needs into technical solutions.

Leads the architecture, development, and technical strategy of Nordstrom’s AI Agentic Platform. Designs agent orchestration, tool use, memory, multi-agent coordination, observability, evaluation, and safety guardrails. Partners with AI/ML and product teams, evaluates LLM and agent frameworks, builds proofs of concept, establishes engineering standards, mentors engineers, resolves complex platform issues, and influences enterprise technology roadmaps.

Lead applied AI model post-training, evaluation, and production ML system development. Adapt models to customer-specific scientific workflows using SFT and reinforcement learning, build evaluation and monitoring systems, debug complex model behavior, collaborate with research and software teams, and mentor engineers. The role requires deep expertise in Python, PyTorch, LLMs, multimodal or agentic systems, and shipping reliable machine learning capabilities.

Own the AI compute software stack for robotic edge devices. Design systems across CPU, GPU, NPU, and FPGA architectures; deploy and optimize neural networks, LLMs, vision models, and deterministic robotics algorithms under real-time constraints. Establish testing and code-quality standards, coordinate cross-functional delivery, plan initiatives, de-risk technical challenges, and maximize hardware performance using optimized C++ and Python. The role requires embedded deployment, hardware execution expertise, accelerator experience, and robotics algorithm knowledge.

Develop and improve robotics R&D infrastructure spanning teleoperation, data annotation, robot manipulation, hardware testing, calibration, and task evaluation. Build software for robot control, inverse kinematics, motion control, CAN communication, and hardware interfaces. Design experiments, validate physical task outcomes, and convert prototypes into reusable, tested, documented tools for AI and production engineering teams.

Build, test, deploy, and operate secure, scalable AI/ML pipelines for national security missions. Develop reproducible workflows, automate operations, integrate LLMs and agentic capabilities, and deploy solutions across hybrid environments. Collaborate with AI engineers, data engineers, cloud architects, and security professionals to optimize data flow, monitoring, observability, infrastructure integration, and mission resilience.

The Senior Software Engineer - AI will design and implement AI solutions for legal workflows, ensure robust system architecture, collaborate with clients, and evaluate new AI techniques. Responsibilities include developing workflows and improving system observability.

Design and implement traditional machine learning and generative AI capabilities for electronic warfare products. Conduct research and development on emerging AI trends, support product design, ensure quality assurance, and transition capabilities into production. The role requires clearance, programming ability, and education or equivalent experience in machine learning or artificial intelligence. Prior intelligence community or Department of War experience is preferred.

Build, integrate, evaluate, and deploy production AI/ML systems across defense product lines. Responsibilities include model serving, data pipelines, inference, monitoring, lifecycle management, robustness testing, interpretability, performance optimization, and edge deployment in constrained or adversarial environments. Collaborate with data scientists, software engineers, product teams, and mission stakeholders to turn research and prototypes into reliable software. The role requires strong Python and production engineering skills, with 10–20% travel and regular on-site work in San Francisco.

Lead a three-person computer vision team developing, training, and deploying custom YOLO object-detection models for live full-motion video. Architect secure video-training pipelines, establish annotation and dataset standards, address class imbalance and challenging environments, and optimize inference using CUDA, TensorRT, and DeepStream. Integrate computer vision capabilities with broader applications, guide technical direction and Agile delivery, and mentor mid-level engineers.

Architect and deploy production-grade generative and agentic AI systems for Citi’s regulated enterprise environment. Responsibilities include context engineering, prompt engineering, RAG, knowledge graphs, Graph RAG, multi-agent orchestration, MCP and A2A integrations, observability, responsible AI controls, scalable deployment, and real-time AI solutions. The role partners with cross-functional stakeholders, mentors engineers, conducts code reviews, and advances automation across Controls Technology.

Own the conversational intelligence layer for an agentic travel platform. Responsibilities include designing agent architecture, prompts, tool calling, state and memory systems, ranking and retrieval, personalized recommendations, streaming, error recovery, model-provider strategy, evaluation systems, and rendering. The role requires production experience with LLM-powered systems, diagnosing reliability issues, and improving quality through user evidence. It is a full-time hybrid position based regularly in New York City.

Build and improve state-of-the-art agent systems: design RL/RLHF learning pipelines, create custom environments, set up robust evaluation, train smaller task-specific models, leverage open-source text and image models, and help shape technical and hiring strategy alongside founders.

Design and implement intelligent automation for AV software development, testing, release, and operational workflows. Build CI/CD pipelines integrating simulation and hardware-in-loop testing, develop engineering data pipelines and metrics, and create AI-enabled tools for test analysis, failure classification, defect triage, reporting, and workflow orchestration. Establish evaluation, governance, security, and monitoring standards for AI solutions while improving release readiness, software quality, engineering productivity, and cross-functional visibility.

Architects and delivers enterprise-scale AI security systems for identity risk, anomalous access, malware detection, sensitive data classification, and threat recognition. Builds inference engines, ML platforms, feature pipelines, and production model-monitoring workflows. Provides technical leadership across multi-quarter initiatives, integrates AI with access-control and security infrastructure, evaluates emerging techniques such as LLMs and agentic systems, and collaborates across identity, threat detection, and platform teams.

Build enterprise applications and polished user interfaces for agentic AI experiences. Own full-stack features from frontend architecture through backend integration, testing, deployment, and production support. Develop React and TypeScript interfaces for multi-step agent workflows, tool calling, approvals, retrieval, state management, errors, and human-in-the-loop interactions. Integrate APIs, agent orchestration layers, and enterprise systems while contributing to backend development, architecture reviews, reliability, and production engineering.

Leads full-stack engineering and AI modernization initiatives, designing scalable enterprise applications and architecture. Responsibilities include coding, testing, technical discovery, application modernization, security and privacy integration, engineering standards, mentoring, stakeholder consulting, and strategic technology leadership. The role requires collaboration across product, operations, data, and security teams while balancing innovation, platform stability, and delivery commitments.

Lead and guide agile teams to design, build, and deliver secure, scalable AI-enabled software. Develop and review production code, drive product and technical decisions, champion enterprise AI-assisted development practices, implement RAG/LLM architectures and vector stores, integrate cloud services and CI/CD, and establish responsible AI guardrails and validation standards.

Build customer integrations, data syncs, AI workflows, internal agents, and automations for permitting operations. Develop evaluations and monitoring to ensure AI reliability, security, and adoption. Work directly with customers, Permit Ops, Product, and Engineering to identify process improvements, turn one-off solutions into reusable tools, and continuously improve production systems.

Embed with customer AI teams to design, prototype, and productionize AI agents using W&B Weave. Advise on architecture, build reference implementations and demos, run technical workshops, gather field feedback for product roadmap, and troubleshoot customer environments.

Design, develop, and productionize LLM and multimodal AI solutions (RAG, agents). Lead internal AI enablement through training, playbooks, office hours, and cross‑functional collaboration to ensure responsible, scalable AI adoption.

Leads and contributes hands-on to development of an AI engineering platform for semiconductor testing. Designs agent tooling, RAG systems, cloud and air-gapped inference infrastructure, Kubernetes platforms, reusable AI patterns, and agent-facing UIs. Sets technical direction, selects hardware and inference stacks, manages production systems, resolves cross-team architecture decisions, and guides a small engineering team. Requires strong distributed systems, software development, leadership, communication, and LLM/agent tooling expertise.

Lead the technical strategy, architecture, modernization, and company-wide adoption of Pinterest’s AI/ML evaluation platform. Design reliable evaluation systems for GenAI and machine learning workloads using automated metrics, human feedback, expert judgment, safety assessments, and business outcomes. Drive cross-team workstreams, resolve technical tradeoffs, establish evaluation standards, and improve the quality, reliability, and trustworthiness of AI-powered products.

Build and ship production generative AI applications, services, agents, and tools using LLM APIs. Develop prompts, structured outputs, tool calling, retrieval, evaluation frameworks, and failure handling. Integrate AI with product APIs and data models, write optimized SQL, deploy cloud services, operate Hugging Face and hosted inference systems, and monitor quality, latency, token usage, and cost. Collaborate with product and engineering teams while contributing tests, code reviews, and internal AI initiatives.

Lead architecture and hands-on development for dexterous robotics software and Physical AI infrastructure. Own manipulation use cases from definition through demonstration, integrating sensing, control, simulation, learned policies, data pipelines, testing, and deployment. Translate ambiguous R&D goals into delivery plans, establish production-grade reliability practices, coordinate cross-functional technical work, and mentor engineers. Develop core software in C++ and Python using ROS 2 while addressing timing, synchronization, safety, reproducibility, and performance constraints.