Distributed ML Systems & AI Platform Engineer

Eric (YunFeng) Li Senior Software Engineer

608-***-**** *******.********@*****.*** Madison, WI LinkedIn SUMMARY

Senior Software Engineer with over 8 years of experience specializing in distributed systems, ML infrastructure, and full-stack AI platform execution. Proven track record architecting exabyte-scale data pipelines for top-tier LLM training clusters (16,000-GPU Meta RSC) alongside deploying autonomous AI agent and RAG frameworks. Combines deep early- stage startup foundations in custom hardware compilation with elite big-tech scaling to build robust, cost-optimized AI systems from the ground up.

PROFESSIONAL EXPERIENCE

Meta Senior Software Engineer Greater Seattle Area Oct 2018 – Apr 2026

• Architected fault-tolerant data pipelines for Meta's 16,000-GPU AI Research SuperCluster (RSC), scaling exabyte-scale training infrastructure for Llama LLMs

• Engineered distributed checkpointing and node recovery systems, increasing training job MTBF by 35% across 16,000 NVIDIA A100 GPUs

• Engineered autonomous AI agent workflows using LangChain, coordinating multi-tool calling to accelerate automated operational task execution by 28%

• Developed context-aware retrieval-augmented generation (RAG) frameworks for Llama models, improving prompt-response alignment metrics by 22% across internal platforms

• Designed high-availability gRPC microservices handling 50k+ QPS, optimizing database query layers to reduce end-to-end API latency by 18%

• Developed high-performance feature pipelines for Andromeda, a personalized ads retrieval engine powering Advantage+ automation for billions of users

• Fine-tuned deep learning recommendation embedding models for Andromeda, driving a 15% increase in click- through-rate precision for ad targeting

• Boosted Andromeda pipeline throughput by 3x via Kafka partitioning optimization, dynamic autoscaling, and low-latency service-level performance tuning

• Deployed containerized ML workloads across hybrid cloud environments using Kubernetes and Terraform, optimizing cloud infrastructure resource utilization by 24%

• Built automated orchestration and validation systems for large-scale ML retrieval workloads, reducing manual production engineering investigations by 45%

• Designed real-time React observability dashboards for pipeline health, cutting production anomaly detection and regression response times by 30%

• Mentored 3 junior engineers through rigorous system design reviews, accelerating team feature delivery velocity and securing 2 promotions

SimpleMachines Inc Software Engineer Madison, WI Sep 2017 – Oct 2018

• Co-designed the compiler toolchain mapping high-level ML graphs to a custom dataflow architecture, accelerating model compilation velocity by 40%

• Developed cycle-accurate architectural simulators to validate chip performance, identifying critical hardware bottlenecks prior to taping out first-generation silicon

• Optimized low-level execution runtimes and custom intrinsic kernels, boosting early-stage tensor operations throughput by 2.5x across simulated environments

TECHNICAL SKILLS

• Languages: Python, C/C++, Java, Go, SQL, TypeScript, JavaScript, Bash/Shell

• Backend Frameworks & Microservices: Spring Boot, FastAPI, Flask, gRPC, Protocol Buffers (Protobuf), RESTful APIs, GraphQL, Node.js/Express.

• AI/ML & Agentic Frameworks: PyTorch, LangChain, LlamaIndex, Hugging Face Transformers, Deep Learning Recommendation Models (DLRM), Triton Inference Server, OpenAI API.

• Vector Databases & Retrieval: Pinecone, Milvus, Qdrant, FAISS, Retrieval-Augmented Generation (RAG) Architecture.

• Distributed Systems & Data Infrastructure: Apache Kafka, Apache Spark, Ray, Distributed Checkpointing, Exabyte-Scale Storage Systems, Redis, Apache Airflow, Hadoop.

• Cloud, DevOps & Orchestration: Kubernetes (K8s), Terraform, Docker, AWS (EC2, S3, EKS), Hybrid Cloud-Native Infrastructure, CI/CD (Jenkins, GitHub Actions).

• Systems Programming & Hardware Architecture: Compiler Toolchains (LLVM concepts), Architectural Simulation, Dataflow Processor Architecture, Low-Level Kernel Optimization, ASIC/Silicon Validation.

• Frontend & Observability: React, Node.js, HTML5/CSS3, Prometheus, Grafana, Real-Time Pipeline Observability, Production Anomaly/Regression Detection.

• Methodologies & Core Competencies: Distributed System Design, Fault-Tolerant Architecture, High-Throughput Data Pipelines, Performance Tuning, Technical Mentorship & Leadership. EDUCATION

University of Wisconsin-Madison 2016 - 2018

Master of Science Computer Science

Zhejiang University 2012 - 2016

Bachelor of Science Information and Computing Science

Contact this candidate