LOGIC BEHIND INNOVATION

AxonLogic is an engineering company in Tashkent building dedicated teams and AI systems for four countries — the systems that sell a country’s train tickets, guard its station gates, and trade markets in under a second.

START A PROJECT INDEX OF WORK ↓ 07

SHEET 02/08 MEASURED, NOT CLAIMED.

0% Client satisfaction

0+ Systems shipped

0 Countries served

0+ Engineers

0 Yrs combined leadership

SHEET 03/08 INDEX OF WORK — fig.01–fig.07

SHIPPED IS THE ONLY PORTFOLIO

Shipped, not slideware — eight production systems in rail, fintech, and AI, running today in four countries.

fig.01 SorbonRAIL · AI Deep reinforcement learning allocating seats across a national rail network. → fig.02 E-TicketRAIL · E-COMMERCE Passenger booking for an entire railway — every device, three languages. → fig.03 Sorbon AnalyticsBUSINESS INTELLIGENCE Live revenue, load and route intelligence for operators. → fig.04 Ticket OfficeRAIL · POS Ticket-office point of sale wired into the same core as online sales. → fig.05 Ticket Validation SystemRAIL · ACCESS CONTROL Station gates that verify every ticket at the turnstile. → fig.06 Sorbon CateringRAIL · SERVICE Onboard food service, ordered and fulfilled per seat. → fig.07 Trade BotDEFI · RUST/GO A copy-trading engine mirroring Solana wallets in under a second. →

fig.01 — SORBON

$ tradebot --watch 8xF3…9kQ
» stream    yellowstone/grpc     LIVE
» decode    pumpswap · raydium   0.4ms
» build     swap (zero-rpc)      1.1ms
» send      jito relay           ✓
── mirrored in 0.03s ────────────────

SHEET 04/08 CASE — SORBON

A NEURAL NETWORK DECIDES WHO SITS WHERE

PROBLEM	Empty seats are decisions made too late.
APPROACH	Deep reinforcement learning re-allocates inventory as demand moves.
RESULT	Higher load factors and revenue across a national network.

READ THE SYSTEM →

EPISODE 41,203 · LOAD FACTOR 94.2% · REVENUE +11.8% (SIMULATION)

SHEET 05/08 SPEC — CAPABILITIES

CAPABILITIES, EACH WITH AN ALIBI

LEFT — WE REASON

Dedicated Teams

Senior engineers embedded in your roadmap — argue, commit, ship.

IN SERVICE: fig.01 SORBON — 4 YEARS RUNNING →

Microservices

Distributed systems that survive national-scale traffic.

IN SERVICE: fig.02 E-TICKET — NATIONWIDE →

Cloud & DevOps

Grafana-instrumented, containerised, observable by default.

IN SERVICE: fig.07 TRADE BOT — 24/7 OPS →

RIGHT — WE FIRE

AI & Machine Learning

Reinforcement learning in production — not a demo notebook.

IN SERVICE: fig.01 SORBON — DRL CORE →

Custom Software

POS, gates, catering, trading — whatever the operation needs.

IN SERVICE: fig.04–06 KASSA · TICKET VALIDATION · CATERING →

Web + Mobile

Every passenger touchpoint, from kiosk to pocket.

IN SERVICE: fig.02 · fig.04 E-TICKET · KASSA →

SHEET 06/08 ENGAGEMENT MODELS — ALL THREE END IN OPERATION

HIRE A TEAM, BUY AN OUTCOME OR FUND THE RESEARCH

01

Dedicated Team

TEAM SIZE	2–10
RAMP-UP	2–3 WEEKS
BILLING	MONTHLY
AFTER LAUNCH	THE TEAM STAYS

Your product, our engineers — a standing team that learns your domain and stays.

DISCUSS THIS MODEL →

02

Fixed-Scope Build

TEAM SIZE	SCOPED
RAMP-UP	1 WEEK
BILLING	MILESTONES
AFTER LAUNCH	WE KEEP IT RUNNING

A defined system, a defined date, a defined price. We carry the risk.

DISCUSS THIS MODEL →

03

AI R&D Partnership

TEAM SIZE	RESEARCH POD
RAMP-UP	IMMEDIATE
BILLING	PHASED
AFTER LAUNCH	THE MODEL KEEPS LEARNING

Applied ML for operations — from feasibility memo to a model in production.

DISCUSS THIS MODEL →

SHEET 07/08 DWG — SYSTEM INTERCONNECTION · fig.01–06

SIX FIGURES, ONE MACHINE

Six of the seven figures in the index are not separate projects — they are one system running a national railway. Follow one passenger tap: booked online or at the window, seat decided by a neural network, verified at the gate, fed at the seat — and the operator watches it all live.

AN AXON IS THE WIRE A SIGNAL TRAVELS DOWN. THE NAME IS NOT DECORATIVE.

NOTE 1 — IN SERVICE 24/7. SCHEDULED HANDOFF — NONE. LONGEST-STANDING TEAM: SORBON, 4 YEARS.

NOTE 2 — DRAWN IN TASHKENT (UTC+5). IN SERVICE ACROSS 4 COUNTRIES.

NOTE 3 — SYSTEMS LIKE THIS ARE THE PROBLEMS EVERYONE ELSE DECLINED. SEE SHEET 08/08 ↓

SHEET 08/08 ISSUE FOR CONSTRUCTION

BRING US THE PROBLEM EVERYONE ELSE DECLINED

START A PROJECT →

REPLY FROM AN ENGINEER — NOT A SALES DECK — WITHIN 24H.