Case study

TenderSRB: an AI Service That Reads Public Tenders

Victor Romanov · Founder and AI Solution Architect · Working product in open beta · Figures on 20 October 2026 · app.nab.rs
60,500+tenders processed
28,700+notifications delivered
7,700+automated tests
34architecture decision records

The task

A supplier who bids on public tenders has to watch the state procurement portal every day.
The portal publishes about 40,000 procedures a year.
Most of them do not fit the company.
The documents are in Serbian, and the technical specification is often a long PDF file.
A tender manager can spend hours on search and still miss the right contract.

What the client gets

Tender matching

Delivery

Work with a tender

AI on Claude models

Public part

Illustration of the client flow. Sample data, not a real tender.

Architecture

VPS · DOCKER COMPOSE · 19 SERVICES State portaleNabavkenew tenders nightly Claude APIcheaper modelstronger model SubscriberTelegram appemail inboxweb browser Stripecheckoutwaits for activation GitHub maintests, AI review Admin bot@tendersrb_bot Ingest workerfetch, normalizepolite rate limit PostgreSQL 16+ pgvector (next step) tenderscompany profilesrow-level securitydelivery outboxAI analyses44 migrations Routing workerA: CPV, region, budget,deadline, procedureB: keyword rules AI workercard translationspec analysis on button Delivery workerTelegram, emailno duplicates, no losses Web cabinet + APIFastAPI, HTMXfeed, funnel, templates Telegram botaiogram 3commands, buttons Caddy 2HTTPSproxy Deployone command, check Redis 7queues, cache, limits Monitoringdaily report, backups tenders notify outbox prompt JSON matched tenders only webhook deploy alerts
Tender data, AI requests, notifications, user requests, payments and operations. Click a color to see one flow.

Five decisions that shaped the system

1. One shared bot for all companies

An early plan was a separate bot for each subscriber.
With many companies, that means many tokens, many webhooks and many points of failure.
So there is one bot, and each company has its own profiles in the database.
A routing worker compares every new tender with all profiles.

2. The database itself keeps client data apart

Every client table has PostgreSQL row-level security rules.
The bot and the API work under a restricted database role.
This role is not a superuser and cannot bypass the rules.
The data of one client stays closed to another client even if the code has a bug.

3. Every stage is a separate service

Intake, normalization, matching, the delivery queue, AI and sending run as separate services, 19 in total.
A failure in one stage does not stop the others.
The delivery queue protects against repeated sending, so there are no duplicates and no losses.
The service does not send to a channel that is known to be broken.

4. AI on request, with a limit check on every call

The client presses a button, and AI reads the specification files.
The answer lists requirements, risks and deadlines, and its depth depends on the plan.
The same buttons give a bid or no bid verdict, a document checklist and a bid draft.
Each call passes a limit check before it starts and goes to the usage records after it ends.
AI cost stays under control, and the client always knows the reason for a refusal.

5. Care for the data source

Requests to the state portal go with a rate limit.
When the portal is overloaded, the intake slows down on its own.
From the first phase, the data model also has a country registry, so a new market needs a new source, not a new product.

How it was built

I am the architect and the owner of the product.
I set the goals, the limits and the budget, give tasks to people and AI agents and accept the result.
A coordinator agent checks that a task has real value in the data, then writes a short brief.
A worker agent writes the code and the tests in an isolated branch and opens a draft pull request.
A second agent reviews the change.
The coordinator runs the quality gate again and checks the live site after release.
Every change passes the linter, the formatter and the full test suite with a coverage threshold.
Every new test must fail when its fix is removed.
High risk changes go only through me: database migrations, deployment, sign in and billing.
751 pull requests were merged this way between 2 May and 20 October 2026.
The service runs on a VPS in Europe, deploys from the main branch with one command and is checked after every deploy.

Three incidents that changed the rules

Public code: github.com/TnSrb1/nabrs
Agent workflow: github.com/TnSrb1/agent-ops

Stack: Python · FastAPI · PostgreSQL 16 + pgvector · Redis 7 · aiogram 3 · Jinja2 + HTMX · Claude API · Docker Compose · Caddy (TLS) · GitHub CI

Status

The platform works on real data in open beta, free during the beta.
By 20 October 2026 it has processed 60,500+ tenders and delivered 28,700+ notifications.
The code base has 1,329 commits since May 2026.
Plans: Free, Pro at 29 euro per month, Business at 79 euro per month, and Enterprise on request.
Stripe payments are built in the code: subscription checkout, webhooks, renewal and expiry emails.
They will go live when the company details are ready.

In development

Watch, match, read, deliver.
The same pattern fits any stream of documents that a team must not miss.

Victor Romanov
Victor Romanov

Presales Solutions Architect. Remote worldwide, open to relocation.