Alexey Sudachen (Principal R&D Engineer - Edge AI + Hybrid Distributed Systems)

Alexey Sudachen

Principal R&D Engineer
Edge AI + Hybrid Distributed Systems | Rust • C++ • Python

Email: alexey@sudachen.name
GitHub: https://github.com/sudachen
LinkedIn: https://linkedin.com/in/sudachen
Time Zone: UTC/GMT-3, Chile

Overview

25+ years of experience bridging the gap between low-level hardware and high-level distributed protocols.

My focus is entirely on technical excellence, system architecture, and core development via the Individual Contributor (IC) track. I bridge the gap between heavy, complex solutions and the harsh resource constraints of local edge environments.

I don’t believe in “perfect architecture” — it doesn’t exist in the real world. My job is to turn technical chaos and business constraints into working systems. Having taken products from a vague idea to the end of their lifecycle, I stick to a pragmatic approach:

I see the “seams”: I know exactly where a system is “crooked” by necessity, and I don’t look the other way.
I understand the context: I realize why the architecture ended up the way it did — whether it was due to tight deadlines, platform limits, or strategic business decisions.
I manage the risks: My main goal is to design systems so these inevitable trade-offs don’t turn into a disaster during scaling or handover to the maintenance team.
I balance technical debt: I am mindful of tech debt and recognize that trying to pay it off completely is not only impossible but often harmful to the project.

My background spans hardware-software co-design (Linux kernel, UEFI-level security) and heavy data workloads (high-load networking, blockchain indexing). Recently, I’ve been applying this systems-level expertise to Deep Tech and Edge AI – integrating Ternary 1.58-bit SLMs and CNNs into Local Agentic Systems built for resource-constrained environments.

CORE EXPERTISE:

Systems: Low-latency architectures, Linux/RTOS kernel, Memory management, Async Rust, C++.
Edge AI & HPC: Ternary 1.58-bit SLMs & CNNs, Local Agentic Systems, Hardware-software co-design.
Distributed Systems: High-performance computing, Hybrid cloud/edge infrastructure.

WHAT I’M OPEN TO: Currently looking for high-impact Principal R&D roles, architectural consulting, and deep-tech advisory where I can solve non-trivial infrastructure challenges. Feel free to connect or drop me a message directly.

Experience

Principal R&D Engineer / Individual Contributor ║ Remote/Consulting — 2025–2026

I provide specialized, short-term software engineering services and part-time consultancy as an independent professional.

I design and implement resilient, low-latency hybrid distributed systems where performance and efficiency are non-negotiable. My goal is to deliver reality-proof architecture that empowers engineering teams and solves non-trivial business challenges.

Open for B2B consulting contracts.

Senior Software Engineer / Founding Engineer ║ Analog One (Remote) — 2022–2025

Architected and led the end-to-end development of a high-performance distributed data ingestion engine with rich analytics GraphQL API in Rust

Owned the system’s architecture from initial prototype to a highly scalable, production-ready solution, making foundational decisions on the technology stack, data models, and API design.
Engineered a high-concurrency event processing system capable of handling 100,000+ data-retrieval triggers per minute across multiple asynchronous, decentralized nodes, utilizing Tokio and async code to minimize latency.
Designed a heterogeneous data-normalization layer using a novel SQL mapping mechanism to unify state from diverse sources, including deterministic virtual machine (VM) logs, remote RPC endpoints, and external REST APIs
Established an engineering culture of reliability by architecting a BDD automated testing framework (Gherkin/Rust), achieving 90% test coverage and implementing strict architectural reviews to ensure long-term maintainability.
Owned the full DevOps lifecycle, implementing CI/CD pipelines via GitHub Actions and managing automated deployments to AWS using Docker and Infrastructure-as-Code principles to ensure 99.9% uptime.

Tools & Technologies: Rust, GraphQL, SQL, WASM, PostgreSQL / Aurora, Ethereum Compatible VMs, ABIs and calldata encoding, Docker, AWS, GitHub Actions, Linux

Senior Software Engineer ║ Spacemesh.io (Remote) — 2020–2022

Contributed to go-spacemesh client development, focusing on math libraries and consensus algorithms to optimize blockchain operations. Maintained CI/CD pipelines and integrated advanced telemetry systems for performance tracking.

Tools & Technologies: Rust, Go, Terraform, Kubernetes, ELK, Google Cloud, Linux, Windows, Docker

Data Scientist II ║ Vacasa (Hybrid) — 2018–2020

Built predictive models using Python and SQL, improving forecasting accuracy and operational efficiency. Automated ETL pipelines and managed infrastructure with Terraform, Docker and AirFlow.

Developed and maintained cloud infrastructure for the team.
Developed a user-friendly image labeling tool for employees to validate/verify labels predicted by the machine learning model.
Participated in developing a model/tool for identifying rental properties from photographs.
Optimized Python code using low-level modules.
Participated in developing ETL pipelines.

Tools & Technologies: Python, SQL, Go, Rust/C++ (python extensions), Java, Scala, Jupyter, Redshift, AirFlow, Terraform, AWS, Linux, Docker

R&D Engineer / C++ Developer ║ Kaspersky Lab / DLP Research (Remote) - 2012-2018

Pioneered solutions for encrypted GPT disk operations and implemented UEFI drivers for smartcards and tokens. Developed an abstraction layer for utilizing Windows DLLs in UEFI environments, enhancing system versatility.

Ported OpenCT and OpenSC libraries to work with smart cards and electronic tokens in UEFI environment.
Implemented special Verdor smart card / tokens protocols to support cryptography functions in UEFI environment.
Created DLL loader with basic implementation of required kernel and user DLLs to use certified Windows cryptographic DLLs in UEFI environment.
Ported several Linux drivers for network cards not supported by UEFI.
Implemented UEFI disk drivers and improved Windows drivers to boot the system from encrypted GPT disk.

Tools & Technologies: R&D, C, C++, Python, x86/x64 Assembly, UEFI, Windows, Drivers, Cryptography, Reverse-Engineering

R&D Engineer / Independent Software Developer ║ Self-Employed — 2006–2012

Founded and ran an independent software studio as the primary engineer. I personally developed all core technologies from scratch while leading a small team to handle routine tasks and customer support.

Selected by Start-Up Chile: Awarded a place in the prestigious global accelerator program, which provided international validation for my technology and business model.

MoleBox Virtualization Solution

Developed and commercialized MoleBox, a lightweight application virtualization suite, selling thousands of copies to customers worldwide.
Engineered a system to create private virtual environments, effectively isolating applications from the underlying operating system and other host software.
Managed the full product lifecycle: from initial architecture and security implementation to global distribution and support.
Current status: The project has been transitioned to Open Source and is available on my GitHub under the GPL license.

SpriteCraft Game Engine

Designed and built SpriteCraft, a multi-language (C++/Python) sprite-based engine tailored for casual game development.
Successfully utilized the engine to develop and launch a series of casual games that achieved strong sales performance on major digital distribution platforms.
Focused on creating a high-performance, flexible framework that simplified the transition from prototyping to a finished commercial product.
Current status: The framework is now open-source and available on my GitHub under the GPL license.

Tools & Technologies: R&D, x86/x64 Assembly, C, C++, Python, Windows Internals, Reverse-Engineering, Virtualization, Game-Development

Team Lead ║ MERA (On-Site) — 2005-2006

I was leading a programming team working on SIPrise VoIP platform. Mainly I was mentoring my co-workers and planning team work.

Tools & Technologies: C#, VoIP (Voice over IP) Low-Latency

Lead Software Engineer ║ Eyeline Communications (On-Site) — 2003-2005

Eyeline USSD/SMS Center.

I participated in development of the SMSC (Short Message Service Center) Core as a core developer. There I also developed SMPP stack and MAP interaction over Ericsson API.

Tools & Technologies: C++, Linux, Solaris, Core Java, Low-Latency, Multi-threading

C++/Java Developer ║ Novosoft Inc. (On-Site) — 2000-2003

I participated in projects as software developer mostly in C++ and Java. For example I developed from scratch first version of Panasonic PlasCoIP. Exactly the part controlling plasma panel. At other project I developed scaling video driver on WinCE mobile devices for dotPocket Software.

Tools & Technologies: C++, Core Java, Linux, Solaris, WinCE, Drivers, Mobile Devices, Low-Latency, Multi-threading