TELKOMSEL · NETWORK DIRECTORATE

Presentasi Terkunci

Materi internal. Masukkan kata sandi untuk membuka.

Kata sandi salah. Coba lagi.
TELKOMSEL · NETWORK DIRECTORATE

AI Agents:
From Chatbots to
Getting Work Done

How autonomous agents actually work, what is already running in production, and how to start

5active agents
27use cases live
Rp0local model cost
02
AGENDA

Five parts, one hour

Short on theory, then most of the time is spent watching the agents work

01

Concept

What an AI agent is, and how it differs from a chatbot

  • 3 slides
  • Five agents working together from a single mini PC
02

Anatomy

Components, tool use, memory, model routing, platforms

  • 5 slides
03

Evidence

The agent fleet and 27 use cases already in production

  • 3 slides
04

Demo

Four live demonstrations in front of the room

  • 4 slides
05

Closing

Security, governance, and how to get started

  • 2 slides
03
CONCEPT

A chatbot answers. An agent delivers.

The difference is not how clever the model is — it is whether the model can act

CHATBOT (e.g. ChatGPT)

AI AGENT (autonomous)

SystemRuns on the provider's cloudRuns on our own hardware
ControlLimited to one chat windowFull — files, network, internal systems
BehaviourReactive: answers when askedAutonomous and proactive: pursues a goal
MemoryUsually limited to one sessionLong-term: recalls context and history
ActionProduces text onlyRuns tools: terminal, files, APIs, schedules
04
ANATOMY

How an agent reaches the outside world

Users talk to it through ordinary messaging apps; behind that sits a gateway, a runtime, tools — and real connections to real services

Telegram

  • WhatsApp
  • Web browser
  • Command line
  • CONNECTED SERVICES
  • Google WorkspaceEmail, calendar meetings, Drive, Sheets
  • OllamaLocal language model, no data leaves

Gateway

  • Control UI (dashboard)
  • live connections the agent uses today — not a roadmap
  • Instagram GraphReads insights, manages posts and reels
  • ComfyUIFour local image models, zero token cost

Telegram

  • WhatsAppSecond channel on the document agent

Agent Runtime

  • What makes it persistent
  • MemoryRemembers preferences, conventions and past work between sessions.
  • Skills85 ready-made capabilities, loaded on demand.
  • Schedules12 jobs run on their own — daily, weekly, monthly.
  • TOOLSTerminal, files, browser, schedules
  • MODELLocal or cloud, chosen per task
  • IsolationEach agent has its own memory, files and permissions.
  • OpenAIImage generation and heavy analysis
  • DeepSeekReasoning engine for four agents
  • X / TwitterPost search and publishing
  • Instagram tokenAuto-refreshed weekly, 60-day expiry

Cloudflare

  • GoBiz / GoFoodCredentials pending — API already researched
05
ANATOMY

The agent loop: perceive, think, act, check

An agent keeps cycling until the goal is met — that repetition is what separates it from a chatbot

01

  • PerceiveReads the request and the current situation
  • If the goal is not met, the loop repeats
  • A REAL RUN, STEP BY STEP
  • Read the fileThe user sends a spreadsheet. The agent opens it and inspects the columns.

02

  • ThinkDecides what the next step should be
  • AnalyseIt groups 159 transactions, computes totals, and checks every subtotal adds up.
03

Produce

04

  • ActCalls a tool or an application
  • VerifyIt opens its own output and re-reads it to confirm nothing is clipped or wrong.

05

  • CheckVerifies whether the goal is actually met
  • Only when met → Result
  • DeliverIt sends the result back to the user in the same chat where the request arrived.
06
ANATOMY

Memory: why an agent never starts from zero

Two kinds of memory work together — the conversation at hand, and permanent notes

Conversation memory

Long-term memory

The session at hand

The message history of the current session, so the agent understands what has just been discussed without being re-briefed.

Permanent notes

Deliberately stored facts: preferences, working conventions, environment details. They survive across sessions and across days.

REAL EXAMPLES FROM OUR SYSTEM

Document archive

The archiving agent files every incoming document into a catalogue holding date, category, summary and source.

Working conventions

The main agent remembers house style: colours, fonts, report layout and where files belong.

Measurable history

Every conversation is stored and can be counted — sessions, messages, even tokens per model.

07
ANATOMY

Model routing: cloud, local, or specialist

This choice decides cost, speed and confidentiality — so we deliberately use all three

01

  • CLOUD MODEL
  • deepseek-v4-flash · 4 of 5 agentsComplex questions, long answers
  • Billed per token, only what is used
  • Best for: content, analysis, heavy reasoning
  • pakisq is the safest agent in the fleetIt runs on a local model AND a locally-hosted agent platform. Nothing is sent to any external service and no data leaves the device — every document is opened, read and processed entirely on our own hardware. That is why confidential documents go to this agent.
  • Split by role

02

  • LOCAL MODEL
  • qwen3.8:27b · 1 of 5 agentsRuns fully on the mini PC we own
  • Zero token cost, no internet needed
  • Best for: archives, private documents
  • Automatic fallback

03

  • SPECIALIST TOOL
  • Claude Code · on demandA dedicated coding and analysis agent
  • Used when the reasoning is genuinely hard
  • Best for: building apps, deep code analysis
  • Configuration, not code
08
ANATOMY

Choosing a platform: four families, many brands

Whatever the brand, the anatomy is the same — a model, tools, memory and a loop

SELF-HOSTED PLATFORMS

  • Runs on your own machine
  • Hermeswhat we run today
  • OpenClawopen-source assistant
  • n8nvisual, self-hostable
  • DifyLLM app platform
  • AutoGPTearly autonomous agent
  • Full control, data never leaves. You own the updates and the uptime.
  • The practical question is not which brand wins — it is whether the data may leave your building, and who controls the tools the agent can reach.

CODING & WORK AGENTS

  • Product you subscribe to
  • Claude Codewe use this too
  • ChatGPT AgentOpenAI
  • Devin AICognition
  • Perplexity Computerresearch agent
  • OpenHandsopen-source coding
  • Deepest at code and long tasks, but the vendor holds the runtime.

ENTERPRISE PLATFORMS

  • Vendor-hosted, governed
  • Microsoft Copilot Studio
  • Google Vertex AI Agent Builder
  • Azure AI Foundrygoverned runtime
  • Salesforce Agentforce 360
  • AWS Bedrock Agents
  • Fastest path inside an existing cloud contract — data leaves the building.

FRAMEWORKS — BUILD IT

  • For developers, in code
  • LangGraphstateful, auditable
  • CrewAIrole-based crews
  • OpenAI Agents SDKGPT-centric
  • Google ADKGemini / Vertex native
  • Microsoft Agent FrameworkAutoGen successor
  • Maximum flexibility, maximum engineering effort — you maintain it.
11
EVIDENCE

What the agents have actually produced

Every tile below is a real artefact from our own machine — screenshots, reports, posters and video

Live web application

  • AI posterText rendered locally, zero cost
  • All of it produced by the agents themselves — including reading their own output to check it before delivery.

Data report

  • pakis
  • Model benchmark4 local image models compared
  • pakissat

Market screening

  • pakis
  • Presentation deckThis deck, built by the agent
  • pakis

Social content

  • pakis
  • Video reelProduced and published to Instagram
  • pakis

pakisem

  • pakisem
12
EVIDENCE

What the agents actually write and publish

Real excerpts from files on our machine, plus video the agents produced end to end

ARCHIVE CATALOGUE

| Date | Category | File | | 2026-08-31 | Finance - Bank | indihome... | | 2026-09-07 | Finance - Invest| lbs-market..| | 2026-09-08 | Family | child-ac-log| Every incoming document filed with a short summary and its source, then searchable.

  • VIDEO OUTPUT — INSTAGRAM REELSThe agent writes the script, lays out every frame, renders the video, and produces the caption and sound track. Four reels are already published to the account.

BUSINESS ACTION PLAN

ONE-MINUTE SUMMARY Two problems fixable this week, at no cost: 1. Not on Google → nobody searching "siomay Margahayu" finds the shop. 2. GoFood rating 4.5 from 7 reviews, and 3 are complaints — below the 4.7 threshold the algorithm favours.

  • pakisq

WEEKLY PERFORMANCE REPORT

7-DAY NUMBERS Reach 227 · Views 860 · Likes 53 · Comments 4 Saves 0 · Shares 0 → saves/reach 0.00% Interaction/reach 25.99% (healthy, but nothing feeds distribution) Reach −26% vs the 30-day average, likes +44% → "liked but not shared" keeps widening.

  • pakissat
  • FOUR REELS PRODUCED SO FARCholesterol — 3.0 MB Insulin resistance — 2.8 MB Dyslipidemia — 4.0 MB Cholesterol & exercise — 3.6 MB Format 9:16. The frames shown here are taken straight from the rendered files.

LIVE ANSWER, 0.4 SECONDS

Q: How many Indomaret stores are there, and which province has the most? A: There are 4,133 Indomaret stores. The province with the highest total number of stores is Jawa Barat (1,520 stores). Answered live on the published dashboard by a model running on Cloudflare.

  • pakisem

pakis

13
DEMO 1

Screening secondary-market issuers

The agent watches the market daily and tells us when something is worth a closer look

What you will see

The agent has already read a 45-page Islamic finance text and turned its investment standards into a checklist.

  • It then scores every issuer on the secondary market against those standards plus price, valuation and liquidity.
  • When an issuer passes the threshold, an alert arrives in chat without being asked.
  • FALLBACK PLANHistorical snapshots and trend charts are already stored, so the demonstration still works if the live feed is unavailable.

How it runs

  • 01Islamic finance rules become explicit screening criteria.
  • Read the standard
  • 02A daily snapshot of prices and demand.
  • Capture data
  • 03Every issuer ranked 0-100 on the same criteria each day.
  • Score
  • 04A short summary plus any issuer worth attention.
  • Alert

Worth noting

The investment framework comes from Ust. Ammi Nur Baits' book on Islamic transactions — real assets, no interest, no uncertainty, no guaranteed returns.

  • The platform itself was founded by and is supervised by Ust. Erwandi Tarmizi, from the same school of Islamic commercial law.
  • The score is a screening aid, not a buy recommendation.
14
DEMO 2

Building an app, then publishing it

The application is built in Claude Code; our agent is what puts it on the internet

LIVE NOW

minimarket-indonesia.pages.dev

9,066 minimarket stores mapped across Indonesia.
Answered 200 OK from anywhere in the world.
Hosting cost: zero. Bandwidth: uncapped.
Stays online even when our mini PC is off.

HOW IT RUNS

01 Build in Claude Code, tested locally 02 Prepare files so no server is needed 03 The agent deploys and verifies data loads 04 A public address is handed back to us

FALLBACK PLAN

The application is already published and reachable, so this demonstration does not depend on the office network or on our mini PC being switched on.

15
DEMO 3

Social content that closes the loop

The agent writes content from the account's own history, then measures what happened

What you will see

The agent drafts carousels and reels based on what has actually been shared most from the account.

  • Once published, it pulls real performance data: reach, saves and shares per post.
  • The audience sees next week's recommendation being built from numbers, not opinion.
  • FALLBACK PLANAll content files and insight reports are stored locally, so the demonstration still runs even if the API quota is exhausted.

How it runs

  • 01Pull performance of previous posts.
  • Read history
  • 02Decide the topic, slide order and text for each slide.
  • Write
  • 03Lay out the slides in the owner's visual identity.
  • Design
  • 04Pull insights and produce the next recommendation.
  • Measure

Worth noting

Health claims are worded carefully so they stay medically safe.

  • Publishing to Instagram is explained verbally rather than demonstrated.
  • Performance data comes from the official API, not from screenshots.
16
DEMO 4

From a hand sketch to a corporate deck

Hand over a sketch on paper; the agent turns it into a finished presentation

What you will see

An audience member is asked to sketch an idea on paper and photograph it.

  • The agent reads the sketch, understands the structure, and builds a deck in the established corporate identity.
  • Charts and diagrams are redrawn as vectors, so everything stays sharp.
  • FALLBACK PLANThe deck currently open in front of the audience is its own proof — nothing needs to be re-generated during the demonstration.

How it runs

  • 01Recognise the intent, order and relationships.
  • Read the sketch
  • 02Lay out the slides and the weight of each section.
  • Structure
  • 03Produce design, charts and illustrations to the identity.
  • Draw
  • 04Verify: no clipped text, no overlapping elements.
  • Check

Worth noting

The deck you are looking at now was made with exactly this flow.

  • One font family and one colour scheme, with no leftovers from foreign templates.
  • Visual quality matters more than the speed of generation.
17
CLOSING

Security and governance

Four rules we have held since day one — every one of them learned from real experience

01

Tokens, never passwords

  • The agent never holds a password. Every integration uses a scoped token that can be revoked without changing anyone's credentials.
  • Permissions limited to what the task needs
  • Stored separately with strict file permissions
  • Short-lived and refreshed automatically
  • THE ONE RULE THAT MATTERS MOST
  • Give an agent access that can be withdrawn, not trust that cannot be recalled.
02

Least privilege

  • Each integration is given the narrowest access that still works. If it only needs to read, it does not get write access.
  • Write access only when genuinely required
  • Each token scoped to a single service
  • Revoking one token does not disturb the rest
03

Agent isolation

  • Every agent lives in its own space: its own memory, files and permissions. One agent cannot read another agent's data.
  • No mixing of working context
  • Conversations do not leak between agents
  • Explicit rules on what must never be disclosed
04

An auditable trail

  • Every conversation, model call and produced artefact is recorded and can be recounted at any time.
  • Sessions, messages and tokens are counted
  • Files archived in a structured catalogue
  • Can trace who asked for what, and when
18
CLOSING

When to use an agent, and how to start

Three small steps to try it without disturbing the work already in flight

01

Thank you

Questions, an idea for a task you would like to try, or a repeat of any demonstration — let us discuss it now.

  • Start with one task
  • Pick a single repetitive task with a clear beginning and a clear result. Do not start from your most complicated process.
  • Example: producing a recurring report from data you already have.
02

Give it a home

You need one machine that can stay switched on — a small computer or a server — plus one language-model account. No special hardware.

  • Example: the mini PC we already use, with nothing added.
03

Q & A

  • Add capabilities
  • Once one task runs reliably, add the next one. Each new capability is stored separately, so nothing already working is disturbed.
  • Example: from producing a report, to watching the market and sending alerts.