SR Principal Software Engineer - LLM Engineering
Koitecc Solutions, Phoenix, AZ, United States
We're looking for a tech leader ready to take their career to new heights. Join the ranks of top talent at one of the world's most influential companies.
Job responsibilities
Advises and leads on the strategy, architecture, and development of Model serving solutions for different model architectures including LLMs & GNNs, across cloud and on‑premises environments, aligning initiatives to business outcomes.
Defines and implements MLOps and LLMOps strategies for end‑to‑end model lifecycle management, including training, versioning, deployment, monitoring, and governance.
Drives optimization of Model inferencing for high throughput and low latency using quantization, model parallelism, intelligent batching, and hardware acceleration for all model architectures.
Creates durable, reusable software and platform frameworks to standardize ML Engineering services, enabling scale across teams and functions.
Establishes best practices for automation, CI/CD, and infrastructure‑as‑code using containerization and orchestration technologies.
Partners closely with data science, platform engineering, and SRE teams to productionize the models on AWS, ensuring observability, reliability, and cost efficiency.
Leads deployment and optimization using Model Inference servers such as Triton Inference Server and vLLM for high‑throughput, low‑latency serving at scale.
Oversees production operations for AI workloads, including monitoring, incident response, security, and compliance, with continuous improvement.
Translates highly complex technical concepts and emerging trends into actionable strategies for executive and product leadership.
Influences senior stakeholders and cross‑functional partners to prioritize and deliver AI/ML capabilities that drive measurable business impact.
Promotes the firm's culture of diversity, opportunity, inclusion, and respect across teams and communities.
Required qualifications, capabilities, and skills
Formal training or certification on software engineering concepts and 10+ years of applied experience.
8+ years of AI/ML engineering experience with significant expertise in LLMs, GNNs and other model architectures (e.g., GPT, Llama, Falcon, Mistral).
Demonstrated success architecting and deploying LLM & GNN solutions on AWS (e.g., SageMaker, Bedrock, EKS) at enterprise scale; experience with Azure ML or GCP Vertex AI.
Experience building LLM, GNN serving platforms in large‑scale environments typical of major tech firms.
Hands‑on experience building LLM inference engines using Triton Inference Server and vLLM, including autoscaling, caching, and throughput optimization.
Advanced proficiency in Python and optimization techniques applied to deep learning frameworks (PyTorch, TensorFlow, Hugging Face Transformers).
Deep understanding of LLMOps/MLOps (e.g., MLflow, SageMaker Pipelines, Kubeflow) with a track record of implementing best practices at scale.
Expertise in inference optimization and distributed systems for large models focused on high‑throughput, low‑latency applications.
Practical experience delivering system design, application development, testing, and operational stability for enterprise AI platforms.
Proven collaboration with SRE to implement observability, incident response, and SLIs/SLOs for LLM services.
Excellent communication skills with the ability to influence both technical and non‑technical stakeholders and deliver value across functions at scale.
Preferred qualifications, capabilities, and skills
Master's or PhD in Computer Science, Engineering, or a related field (or equivalent experience).
Practical cloud‑native experience, including containerization (Docker), orchestration (Kubernetes), and infrastructure‑as‑code (Terraform, CloudFormation).
Expertise in security, compliance, and governance for AI/ML deployments in regulated environments.
Experience in trust and safety or fraud prevention domains; familiarity with payments platforms is a plus.
Track record of contributions to open‑source LLM projects or peer‑reviewed research and/or experience presenting at industry conferences or leading technical communities.
Familiarity with hardware acceleration strategies across GPUs, TPUs, and specialized inference runtimes.
Experience in building java based applications.
This position is subject to Section 19 of the Federal Deposit Insurance Act. As such, an employment offer for this position is contingent on JPMorgan Chase's review of criminal conviction history, including pretrial diversions or program entries.
We are an equal opportunity employer and place a high value on diversity and inclusion at our company. We do not discriminate on the basis of any protected attribute, including race, religion, color, national origin, gender, sexual orientation, gender identity, gender expression, age, marital or veteran status, pregnancy or disability, or any other basis protected under applicable law. We also make reasonable accommodations for applicants' and employees' religious practices and beliefs, as well as mental health or physical disability needs. Visit our FAQs for more information about requesting an accommodation.
#J-18808-Ljbffr
Job responsibilities
Advises and leads on the strategy, architecture, and development of Model serving solutions for different model architectures including LLMs & GNNs, across cloud and on‑premises environments, aligning initiatives to business outcomes.
Defines and implements MLOps and LLMOps strategies for end‑to‑end model lifecycle management, including training, versioning, deployment, monitoring, and governance.
Drives optimization of Model inferencing for high throughput and low latency using quantization, model parallelism, intelligent batching, and hardware acceleration for all model architectures.
Creates durable, reusable software and platform frameworks to standardize ML Engineering services, enabling scale across teams and functions.
Establishes best practices for automation, CI/CD, and infrastructure‑as‑code using containerization and orchestration technologies.
Partners closely with data science, platform engineering, and SRE teams to productionize the models on AWS, ensuring observability, reliability, and cost efficiency.
Leads deployment and optimization using Model Inference servers such as Triton Inference Server and vLLM for high‑throughput, low‑latency serving at scale.
Oversees production operations for AI workloads, including monitoring, incident response, security, and compliance, with continuous improvement.
Translates highly complex technical concepts and emerging trends into actionable strategies for executive and product leadership.
Influences senior stakeholders and cross‑functional partners to prioritize and deliver AI/ML capabilities that drive measurable business impact.
Promotes the firm's culture of diversity, opportunity, inclusion, and respect across teams and communities.
Required qualifications, capabilities, and skills
Formal training or certification on software engineering concepts and 10+ years of applied experience.
8+ years of AI/ML engineering experience with significant expertise in LLMs, GNNs and other model architectures (e.g., GPT, Llama, Falcon, Mistral).
Demonstrated success architecting and deploying LLM & GNN solutions on AWS (e.g., SageMaker, Bedrock, EKS) at enterprise scale; experience with Azure ML or GCP Vertex AI.
Experience building LLM, GNN serving platforms in large‑scale environments typical of major tech firms.
Hands‑on experience building LLM inference engines using Triton Inference Server and vLLM, including autoscaling, caching, and throughput optimization.
Advanced proficiency in Python and optimization techniques applied to deep learning frameworks (PyTorch, TensorFlow, Hugging Face Transformers).
Deep understanding of LLMOps/MLOps (e.g., MLflow, SageMaker Pipelines, Kubeflow) with a track record of implementing best practices at scale.
Expertise in inference optimization and distributed systems for large models focused on high‑throughput, low‑latency applications.
Practical experience delivering system design, application development, testing, and operational stability for enterprise AI platforms.
Proven collaboration with SRE to implement observability, incident response, and SLIs/SLOs for LLM services.
Excellent communication skills with the ability to influence both technical and non‑technical stakeholders and deliver value across functions at scale.
Preferred qualifications, capabilities, and skills
Master's or PhD in Computer Science, Engineering, or a related field (or equivalent experience).
Practical cloud‑native experience, including containerization (Docker), orchestration (Kubernetes), and infrastructure‑as‑code (Terraform, CloudFormation).
Expertise in security, compliance, and governance for AI/ML deployments in regulated environments.
Experience in trust and safety or fraud prevention domains; familiarity with payments platforms is a plus.
Track record of contributions to open‑source LLM projects or peer‑reviewed research and/or experience presenting at industry conferences or leading technical communities.
Familiarity with hardware acceleration strategies across GPUs, TPUs, and specialized inference runtimes.
Experience in building java based applications.
This position is subject to Section 19 of the Federal Deposit Insurance Act. As such, an employment offer for this position is contingent on JPMorgan Chase's review of criminal conviction history, including pretrial diversions or program entries.
We are an equal opportunity employer and place a high value on diversity and inclusion at our company. We do not discriminate on the basis of any protected attribute, including race, religion, color, national origin, gender, sexual orientation, gender identity, gender expression, age, marital or veteran status, pregnancy or disability, or any other basis protected under applicable law. We also make reasonable accommodations for applicants' and employees' religious practices and beliefs, as well as mental health or physical disability needs. Visit our FAQs for more information about requesting an accommodation.
#J-18808-Ljbffr