Adam Adli
Senior Software Engineer (Senior Mts) @Cerebras
Signup · Get unlimited contacts
WORK HISTORY
Senior Software Engineer (Senior Mts) @Cerebras
Toronto, ON, CA
As Tech Lead, architected and drove the development of Cerebras’ core inference testing infrastructure from the ground up, delivering transformational systems-level solutions that redefined scalable ML product validation and unlocked enterprise-grade performance and developer velocity across the organization: Architected a unified execution-context library for local and remote operations, incorporating advanced metaprogramming, type-safe APIs, and fault-tolerant distributed coordination: Solved core pytest-xdist limitations by- Enabling true session-scoped resource sharing across distributed workers- Creating advanced resource orchestration with flexible synchronization patterns- Developing scheduling algorithms for optimal SLURM cluster utilization: Applied advanced design patterns to deliver modular, extensible frameworks that amplified engineering impact across teams: Delivered significant speed and efficiency improvements in existing infrastructure and benchmarking workflows using optimized parallel programming, networking, and Python asyncio.
EDUCATION
University of Toronto
Honours Bachelor of Science (H.B.Sc.), Computer Science
University of Toronto
Master's degree, Computer Science
ABOUT ADAM ADLI
Tech Lead and Senior Software Engineer at Cerebras with 6+ years of experience building ML infrastructure at scale. I architect the infrastructure that powers AI products: distributed testing frameworks, orchestration systems, and validation pipelines.Expertise includes- AI infrastructure: distributed test frameworks, multi-cluster orchestration, release validation pipelines- Python ecosystems: asyncio, metaprogramming, pytest internals, Django- Distributed systems: HPC cluster scheduling, high-performance parallel programming, profiling- Data & backends: PostgreSQL, real-time monitoring, API designAt Cerebras, I lead Core Infrastructure for Inference - building a multi-cluster scheduling platform for intelligent workload routing and solving hard problems in distributed test execution that remain unaddressed in open source. I\'ve architected pytest plugins that scaled ML validation and dramatically improved performance and developer velocity.At Intel (Altera), I pioneered LLM-based solutions before they were mainstream, including a RAG pipeline that cut document processing time by 95% and an internal chatbot that drove company-wide AI adoption.I see software engineering as both craft and architecture: the art of building elegant abstractions that make complex systems not only possible, but impactful. I care about developer velocity, clean abstractions, and leaving codebases better than I found them.
This profile is compiled from publicly available professional sources. Unifers is not affiliated with or endorsed by LinkedIn. Request removal of this profile.