AGENT.RUNTIME · v3.0.0 · build #2026.04.27 aleksbuss/portfolio
0%
handshake
aleksbuss v3.0
cpu0.40ms tz— CET langen/ru/lv/de statusONLINE live8 systems
Contact C Hire → H
№ 2026 / FOLIO
Available now
Agent [00:00.812] handshake authenticated · session opened · greeting human

AI engineer · multi-agent systems.
Production systems, shipped solo.
Based Hof / Saale, DE

Status EU citizen · remote-first
Credentials DEKRA №2023337
Tel-Ran Berlin · 960h
Stack TypeScript · Next.js · MoA
CF Workers · Whisper · Ollama
View work → CV
mixture-of-agents ● typescript · next.js 15 ● cloudflare workers ● whisper · llama · gemini ● telegram mini-apps ● vitest · playwright ● docker · systemd ● hmac-sha256 ● ollama · local llm ● n8n agents ● supabase · firebase ●
§ I — At a glance
01/06

A short ledger of what shipped.

Every figure points to live infrastructure — production-grade systems I architected, built, and deployed end-to-end under my own name.

i.
8
Production systems
All deployed & live
ii.
2600+
Tests in the flagship
Orchestra · Vitest + Playwright
iii.
0mo
Zero to eight live
Solo delivery, since 07/2025
iv.
0
Languages spoken
RU · LV · EN · DE
§ II — The route
02/06

Fourteen years in logistics.
Now I move features into production.

I didn't take the conventional route. I took the self-disciplined one.

Fourteen years in multicultural logistics — Tesco Carlisle, long-haul freight across the UK and Europe — taught me systems, processes, and how to deliver under pressure.

In 2022 I retrained as a Frontend Developer at Tel-Ran Berlin (960h, DEKRA-certified). Then I went deeper: LLMs, prompt engineering, serverless, local AI. Nights and weekends, while driving part-time to pay the bills.

From mid-2025, I've been shipping. Not learning. Real systems, real users, real infrastructure — under my own name.

Self-discipline is the skill that transfers. Everything else I learn fast — and I have.
2008 — 22
Logistics, UK & Germany
Tesco Carlisle (WMS, planograms), long-haul freight. Process, resilience, multicultural teams.
10/22 — 10/23
Frontend Developer · Tel-Ran Berlin
DEKRA-certified, 960h. JS Advanced, React, DSA, DevOps, QA, UI/UX.
2023 — 25
Deep self-study in AI & automation
n8n, Cloudflare Workers, Ollama, prompt engineering, local LLM, Docker.
07/25 — now
Production mode · 8 live systems
End-to-end: architecture → deployment → monitoring. Solo delivery, real users.
Now
Open to your team
Immediate availability. Germany-based, EU citizen. Remote or on-site.
§ III — Approach
03/06

What an early-stage team needs
— in one hire.

Startups don't need specialists. They need operators who ship features end-to-end without waiting for permission. Six things I bring on day one.

i. Ship rate

Founder-grade ownership.

Spec → architecture → code → deploy → monitor → iterate. No hand-offs, no "not my job". Eight systems shipped solo since mid-2025. The cadence your runway needs.

ii. Full stack

AI, backend & infra in one.

TypeScript/Next.js, LLM pipelines, multi-agent orchestration, FastAPI/Python, Cloudflare edge, Docker/systemd. One headcount, full stack — your seed runway stretches further.

iii. AI-native

LLM is native, not a bolt-on.

I build agent systems, not wrappers: Mixture-of-Agents pipelines, disagreement detection, reflection loops. Multi-provider failover (OpenAI / Gemini / Groq / Ollama), structured outputs, voice pipelines. Designed with latency, cost & failure modes in mind.

iv. Production hygiene

Defaults, not afterthoughts.

HMAC-SHA256, zero-secrets architecture, rate limiting, systemd auto-restart, health checks, webhook verification, secrets via env. Built in from day one.

v. Late-career grit

Self-driven, under pressure.

Career-switched at 35+ and shipped eight systems solo while moving from part-time driving into full-time engineering. Not talent — discipline. I don't need reminders to unblock myself.

vi. Full product

Beyond the backend.

I can design, implement, and launch an AI product end to end — voice, visual generation (ComfyUI, Stable Diffusion), UI, copy, deployment. Useful when your team is three people and a deadline.

§ IV — Work · Selected
04/06

Production systems,
shipped solo.

A selection from eight systems below — each deployed on infrastructure I own and operate, architected and built end-to-end to production standards.

01 05 / 2026 — ongoing Flagship Open Source

Orchestra — Multi-Agent AI Workspace

Mixture-of-Agents pipeline · code-guaranteed Skeptic · live cost telemetry.
Self-hosted AI workspace that runs 3–5 specialised expert agents in parallel on every substantive turn. Embedding-based disagreement detection surfaces expert conflict instead of smoothing it away; a reflection critic revises the final answer. 2,600+ tests, 70+ documented post-mortems, CI — engineering-led, MIT-licensed.
Router (DPG)→proposers ∥ Skeptic→disagreement detector→Aggregator→reflection
Read case study →
live
2600+tests
70+post-mortems
5parallel agents
MITopen source
TypeScript Next.js 15 MoA Vitest Playwright
02 08 / 2025 — ongoing Production Live

AI Dictaphone v7.0

Edge voice pipeline · Whisper → LLM · sub-second latency.
Serverless voice-to-insight pipeline on Cloudflare edge. User voice note → Whisper transcription → LLM summarization → rich Mini-App UI. Zero cold-start, global edge latency, zero secrets leaked to client.
Telegram Mini-App→Cloudflare Workers→Groq Whisper + LLaMA 3.3 70B→response
Read case study →
live
Livein production
0 mscold start
HMACwebhook auth
Edgeglobal
Cloudflare Workers Groq Whisper LLaMA 3.3 JS ESM
03 09 / 2025 — ongoing Live

AI Moderation Bot

n8n workflow · dual-AI failover · auto-escalation.
Production AI agent workflow for Telegram moderation. 15+ node n8n pipeline classifies every message with structured JSON outputs, routes to action, escalates sanctions. Dual-provider failover keeps it running when any LLM API fails.
Message→Filter→Nemotron→fallback GPT-4.1→Router→Telegram
Read case study →
live
15+n8n nodes
2AI providers
3-tiersanctions
JSONstructured
n8n OpenRouter GPT-4.1-mini Telegram API
04 10 / 2025 — ongoing SaaS product Live

AI Psychology Bot

Full SaaS · payments · crisis support · CBT exercises.
End-to-end SaaS built solo to prove out full payment-system integration: distributed serverless backend + auth + Telegram Stars payments + atomic credits + referral program + AI failover. The hard parts wired cleanly, end to end.
Mini-App→HMAC-SHA256→Apps Script→Firebase→Gemini + TTS
Read case study →
live
7-skillprotocol
Starspayments
Atomiccredit ops
Livedeployed
Gemini 2.5 Telegram Stars Firebase Apps Script
05 02 / 2025 — ongoing bushmark.cc Live

Bushmark — Secure Clipboard

Self-hosted · cookie auth · one-click install.
Self-hosted secure clipboard solving a real daily problem: transferring text between servers and devices without USB or third-party services. Three persistent slots, 24h history, auto-rotation, Tailwind UI.
Client→Nginx reverse proxy→FastAPI (systemd)→disk + 30d rotation
Read case study →
live
1-clickbash install
24.04Ubuntu VPS
24hhistory
3slots
FastAPI Nginx systemd Tailwind
06 07 / 2025 — ongoing Self-hosted Live

CyberDed Ultra + ComfyUI

Local AI infra · one-click installer · Android + Linux.
One-click installer for local AI infrastructure. Flask backend with Server-Sent Events streaming chat, real-time system monitoring, Docker-Compose ComfyUI stack with NVIDIA auto-config. Full bash automation.
Bash installer→Ollama + ComfyUI Docker→Flask (SSE)→NVIDIA Toolkit
Read case study →
live
SSEstreaming
GPUauto-detect
Localself-hosted
2 OSLinux + Android
Docker Ollama Flask ComfyUI
07 11 / 2025 — ongoing Open Source Live

Termux AI Dictaphone

Android as offline AI server · zero cloud cost.
Local AI deployment product: complete offline AI stack on any Android phone — Whisper transcription + LLaMA/Gemma inference running on-device. Zero cloud cost, full privacy. Published OSS.
Termux (Android)→faster-whisper / whisper.cpp→Ollama (LLaMA, Gemma)
Read case study →
live
100%offline
$ 0API cost
Dualmode
On-deviceprivacy
Termux faster-whisper Ollama Python
Explore all repositories on GitHub →
§ V — Toolkit
05/06

Full-stack AI engineering,
top to bottom.

From edge serverless to local LLM on Android — every tool here I've used in production.

i. AI · LLM · Agents
Mixture-of-Agents (MoA) Multi-agent orchestration OpenAI GPT-4 / 4.1 / TTS Google Gemini 2.5 Ollama · LLaMA · Gemma Disagreement detection · embeddings Groq Whisper-large-v3 OpenRouter · multi-provider failover Prompt engineering Structured outputs RAG-ready patterns faster-whisper · whisper.cpp
ii. Automation & No-code
n8n · workflows · agents Webhooks & HTTP routing Code Nodes (JS/Python) Error handling & retries Chain-of-Thought architecture Multi-protocol agents
iii. Programming
TypeScript · strict Next.js 15 · React JavaScript · ESM · async Python · Flask · FastAPI Vitest · Playwright Bash · systems scripting HTML5 · CSS3 · responsive Streaming APIs (SSE)
iv. Cloud & Serverless
Cloudflare Workers · edge Google Apps Script Firebase Realtime DB GitHub Pages Supabase / PostgreSQL
v. Production & Ops
Docker · Compose Nginx · reverse proxy Ubuntu VPS · systemd Health checks · auto-restart Secrets · env vars Git/GitHub · Wrangler NVIDIA Container Toolkit Background jobs · rotation
vi. Security & Payments
HMAC-SHA256 · JWT Telegram Bot API · Mini-Apps Telegram Stars CORS · secrets mgmt Anti-fraud logic Rate limiting · webhook verify
RU
Russian
Native
LV
Latvian
C2 · Fluent
EN
English
C1 · Business
DE
German
B1
§ VI — Now hiring myself out

Let's turn your backlog
into shipped software.

I'm looking for AI Engineer or Founding Engineer roles at early-stage startups (Seed–Series C). I build multi-agent systems, write production code, own features end-to-end, and ship fast. EU citizen, based in Germany, available immediately.

Request an interview → See the code Download CV
Email
aleksbuss@gmail.com
GitHub
/aleksbuss
LinkedIn
/aleksejs-buss
Location
Hof / Saale · DE