SYS_STATUS: ACTIVE|DISCIPLINE: DISTRIBUTED SYSTEMS & AI RUNTIME|LOC: Vancouver, Canada / Remote

Michael Lipski

Full Stack Software Engineer

I'm a backend-focused engineer who loves untangling complex distributed systems. I have a proven track record scaling enterprise platforms and a deep curiosity for how we can better architect emerging AI infrastructure.

When I'm not explicitly designing systems, I'm usually ice dancing, playing piano, or digging into a good fantasy novel — exploring non-technical systems to draw inspiration from.

View Systems Architecture ↓Launch Axioma Governance Demo↗GitHub ↗LinkedIn ↗

01 //

Core Competencies & Technical Matrix

A targeted operational surface focused on high-throughput backend services, fault-tolerant infrastructure, and zero-trust runtime boundaries.

Languages

[4 ACTIVE]

▹Go
▹Python
▹TypeScript / JavaScript
▹SQL

Infrastructure

[5 ACTIVE]

▹Kubernetes (GKE)
▹Docker
▹CI/CD
▹RabbitMQ
▹PostgreSQL

Domains

[4 ACTIVE]

▹Distributed Systems
▹API Design
▹Systems Modeling
▹AI Evaluation Context

02 //

R&D & Systems Architecture — Musings Turned Experiments

Exploratory research systems formalizing non-deterministic LLM execution into deterministic state automata and recursive axiomatic compilers.

Axioma

// Dual-FSM Zero-Trust Governance Kernel

Launch Interactive Demo (/axioma)↗

A zero-trust LLM governance architecture utilizing dual finite state machine boundaries to enforce secure, constrained agentic execution. Built with Next.js and TypeScript.

State IsolationDual FSM Enclaves

Deterministic Halt0 Clock Cycles

SubstrateNext.js / TypeScript

Next.jsTypeScriptFinite State MachinesZero-TrustLLM SecurityMulti-Zone Route: /axioma/:path*

Meta-Axioma

// Recursive Axiomatic Compiler & Rule Registry

A Python-based evaluation engine for researching and constructing self-assembling software axioms, featuring strict human-in-the-loop oversight and persistent rule registries.

Axiom Registry16 Formal Invariants

Decay ModelThermodynamic Landauer

Verification GateAnti-Hubris Enclave

Python 3.14AsyncIOCategory TheoryHITL OraclesRule Registries

03 //

Professional Experience (Abridged)

Proven history scaling distributed backend microservices, high-volume ingestion queues, and enterprise SaaS customer touchpoints.

Beehive AI, Inc.

Full Stack Software Engineer//June 2023 – April 2026

View Overseen Systems Stack ↓

Early engineer at a venture-backed AI SaaS startup driving platform scalability, enterprise integrations, and LLM evaluation tooling; architected high-throughput microservices on GKE supporting 10M+ customer records.

—AI Evaluation Engine: Engineered an end-to-end automated LLM evaluation and QA benchmarking system across Go and React, accelerating audit turnaround by ~80% through automated F1 scoring, rate-limited request throttling, and an interactive review dashboard.

—Pipeline Scaling: Re-architected batch ingestion to scale from 10k to 1M+ records per batch, preventing out-of-memory (OOM) failures by designing a dynamic row/size chunking engine to parallelize workloads across a Go worker mesh using streaming row-write flushing, gRPC / Protobuf, and PostgreSQL.

—Enterprise Connectors: Built an end-to-end event-driven integration framework syncing data in near-real-time from Zendesk and Medallia, combining a React OAuth/credential suite with a Go webhook receiver, RabbitMQ queues, and resilient consumer retry policies.

—Clustering Infrastructure: Scaled taxonomy curation by building the distributed K8s worker infrastructure and admin curation tooling to automate detection, clustering, and deduplication of thousands of fragmented/orphaned labels across multi-tenant dashboards.

—Data Governance & SOC 2: Fortified platform security and tenant isolation for SOC 2 readiness across 10M+ records by implementing automated data sanitization jobs, multi-tenant privacy controls, granular audit logging, and scheduled ingestion health alerting.

Stack:GoReactPostgreSQLKubernetes (GKE)gRPC / ProtobufRabbitMQDocker

Globality, Inc.

Software Engineer II//May 2022 – April 2023

Full stack engineer at a Series D AI-driven B2B procurement platform; engineered enterprise workflows using React, GraphQL, and Python microservices deployed across AWS ECS/EKS.

—Technical Lead: Designed and delivered a key customer-facing full stack feature to facilitate competition between providers, aligning with product, design, and QA to communicate realistic expectations and a successful implementation plan.

—Reliability & Observability: Hardened production stability by managing CI/CD lifecycles through Harness and participating in deep-dive post-mortems to identify systemic failure modes and implement preventative logic.

—Data Modeling for Insights: Architected complex PostgreSQL schemas to capture granular user-interaction telemetry, transforming raw event data into actionable datasets for strategic product modeling.

Stack:ReactGraphQLPythonPostgreSQLAWS ECS/EKSHarness CI/CDDocker

04 //

Beehive AI: Overseen Systems Architecture

CORE SCOPE HIGHLIGHTED

Interactive architectural graph of the high-throughput platform at Beehive AI. Everything highlighted in emerald represents the direct domain of ownership—specifically architecting the high-throughput Go APIs, the exploded ETL ingestion pipeline, real-time webhook ingress, interactive AI microservices, and the PostgreSQL partition matrix scaling past 10 million records.

Trace Flow:

Direct Domain of Ownership|Partner ML Team Domain|Client / External Ingress💡 Hover/Click nodes to trace live data vectors

PIPELINE

ETL: Go Normalization Mesh

Role: Transformation & Dedup Core | Tech: Go Goroutine Pools

★ Direct Domain of Ownership

Concurrent worker pools performing schema validation, deduplication, and ML enrichment dispatch.

Key Impact Metrics:Goroutine worker poolsDeduplication & validationGraceful backpressure

Upstream Sources:

Downstream Consumers: