Open to opportunities

Kirill Kuklin

@kirillkuklin

Message

Generative AI engineer specializing in production LLM deployment, RAG pipelines, and low-latency scalable inference.

Canada

Message

What I'm looking for

I’m looking for roles where I can build and operate production GenAI—LLM deployment, RAG, and low-latency inference—while owning MLOps reliability, evaluation, and automation to keep uptime high and deployments fast.

I’m a Generative AI Engineer with 12 years of experience designing and deploying production AI systems, from LLM inference optimization to scalable GenAI infrastructure. I focus on Python-based model serving, LLM deployment and fine-tuning, and end-to-end production ML operations.

At Microsoft, I design and deploy LLM-powered generative AI applications, optimizing inference for production scale. I reduced p99 serving latency by 42% using quantization and batching strategies, and I architected distributed serving handling millions of requests per day with 99.95% uptime.

I also build knowledge-grounded RAG pipelines and evaluation frameworks, fine-tuning and validating foundation models against production baselines before deployment. Previously at Veeam, I improved throughput by 3x and reduced inference latency using serving optimization, and I’ve automated deployment pipelines to cut cycle time from 3 days to 4 hours while maintaining 24/7 reliability.

Experience

Work history, roles, and key accomplishments

Current

Generative AI Engineer

Current

Microsoft

Jun 2021 - Present (5 years 1 month)

Designed and deployed generative AI applications with LLMs, optimizing inference for production scale and reducing p99 serving latency. Architected scalable GenAI infrastructure with RAG pipelines, fine-tuning validation, AI agents, and automated deployment pipelines.

LLM Inference Optimization Vertex AI Fault Tolerance Load Balancing Streaming Pipelines Retrieval Quality Evaluation Generation Accuracy Evaluation Fine Tuning Model Quality Testing Frameworks Regression Benchmarks Python Based Model Quality Testing AI Agents Cloud Build Infrastructure as Code Streaming Brokers Cloud Serving Endpoints

Generative AI Engineer

Veeam

May 2014 - Jun 2021 (7 years 1 month)

Deployed and operated Python-based generative AI models in production cloud environments supporting 24/7 uptime. Built RAG and knowledge retrieval systems, improved LLM serving throughput and latency, and integrated ML scoring outputs into backend services.

Generative AI Model Quantization Throughput REST Adapters For Synchronous Inference Queue Based Adapters For Asynchronous Batch Scoring Capacity Planning Load Testing Model Monitoring Mentoring Engineers

DevOps Engineer

Crossover

Nov 2015 - May 2016 (6 months)

Supported production generative AI pipelines across AWS and GCP by troubleshooting LLM serving failures and inference latency spikes. Developed Python automation for orchestration and artifact management and implemented model monitoring and alerting patterns to detect prediction drift and serving degradation.

AWS GCP Inference Latency Monitoring Model Artifact Management Model Monitoring Root Cause Analysis