Try it on your own work, in your own job
AI is new, and most people in freight have heard a great deal about it and understood very little. Reading about it only gets you so far. So this is not a course. It is a place to put your hands on the thing.
Three programmes, one for each role in a forwarding business. Every one is one to one. You apply, a person reads it and calls you, and if it fits we set up your own space in FreighAI with a slice of your own work in it and walk you through it.
- AI Impact AssessmentOwners, directors and CXOs · Assessment · One to oneBy applicationApply
- Quotation TAT MasterclassPricing heads and managers · Masterclass · One to oneBy applicationApply
- Pricing Desk WorkshopPricing desk staff · Workshop · One to oneBy applicationApply
- Quotation TAT Reduction ProgrammeOwner and head of ops · The fourth step · 3 monthsBy applicationApply
- The course: what AI changes, in six lessonsAnyone · Course · About 75 minOpenOpen now
- The toolkit: twelve questions to ask any vendorAnyone · One sheet · Print itOpenOpen now
- AI in FreightAnyone in freight · Webinar · 45 minNext dateSat 24 Oct
Three programmes. Which one is yours?
Different jobs want different things and are worried about different things. Read the three lines below. Two of them belong to somebody else.
The gap is not ports and roads. It is a Tuesday on a pricing desk.
We could not find another place where a forwarder can put their hands on agentic AI, on their own work, without being sold to. So we built one.
Agentic AI is new, and the noise around it is loud. Most people in freight have heard a great deal about it and understood very little, which is a bad position from which to spend money — or to decide not to.
Meanwhile the volume this industry moves runs on processes that live in inboxes, spreadsheets and people’s heads. That is not a criticism of anybody. It is what the tools allowed until recently.
India’s National Logistics Policy, launched in September 2022, set the country a target: among the top 25 in the World Bank’s Logistics Performance Index by 2030, and logistics costs down towards single digits as a share of GDP. Ports, roads and rail carry most of that. But some of it is decided on a pricing desk on an ordinary Tuesday, in the hours between an enquiry landing and a quote going out.
That is the part we can do something about, and it is why every one of these is free and why nobody is asked to buy. The point is that the people who actually run these desks understand what agentic AI does, and what it does not, before somebody sells them something.
From applying to sitting in your own space
Five steps. Nothing is booked until we have spoken to you, and you hear back either way.
- 01You apply. A short form. Your role, what your work looks like today, and the one thing you would want to see change. Ten minutes at most.
- 02A person reads it. Not a filter and not a scoring model. Somebody from our team reads what you wrote and works out whether this programme is the right one for your role.
- 03We call you. A short conversation, so we understand your desk before we build anything. You can ask us anything here, including what it costs and whether you should be doing this at all.
- 04We sign it off, or we say so. If it fits, we confirm and book the session with you. If it does not, we say that plainly and point you at what would help instead. Either way you hear back.
- 05You get your own space, with your own work in it. We set up an environment in FreighAI for you, with a slice of your real work loaded into it. One of our people sits with you and walks you through it, step by step, until you can drive it yourself.
We are a small team and every session is run by a person, on your own work. So we read each application and talk to you first. It is not about keeping people out. It is about not wasting your morning on something that was never going to fit.
One to one means one company at a time, not one person. Bring anyone else from your firm who should be there, as long as this is the right programme for their job. You are never in a room with another company.
You get your own space in FreighAI, with a slice of your own work in it, and a person walking you through it. Not a recorded demo and not somebody else's sample data.
The session is about your work, not our slides. You will see FreighAI, because that is the thing you are putting your hands on, and at the end we will tell you plainly what is free and what is paid. Nobody is obliged to buy anything.
Here is what it actually changes
Most of what a forwarding office does all day is not judgement. It is reading, fetching, typing, chasing and passing work along. That is the part AI now does.
| It... | Which means |
|---|---|
| Reads | Opens the enquiry, the PDF, the Excel, the WhatsApp photo. Works out what is being asked. |
| Fetches | Pulls your rate cards. Checks the carrier. Looks up what that customer paid before. Reads the shipment status. Talks to your TMS and your accounts system. |
| Drafts | Writes the reply, the request to agents, the counter-offer, the tracking answer, the task for the next desk. |
| Chases | Follows up agents, carriers and customers on its own. Expires and revives quotes. Nobody has to remember. |
| Hands over | Passes the work from pricing to operations to accounts with everything attached. No forwarding. No re-typing. |
- Reads
- Which means
- Opens the enquiry, the PDF, the Excel, the WhatsApp photo. Works out what is being asked.
- Fetches
- Which means
- Pulls your rate cards. Checks the carrier. Looks up what that customer paid before. Reads the shipment status. Talks to your TMS and your accounts system.
- Drafts
- Which means
- Writes the reply, the request to agents, the counter-offer, the tracking answer, the task for the next desk.
- Chases
- Which means
- Follows up agents, carriers and customers on its own. Expires and revives quotes. Nobody has to remember.
- Hands over
- Which means
- Passes the work from pricing to operations to accounts with everything attached. No forwarding. No re-typing.
Your team approves every reply before it sends. Every message a customer sees waits for a person. The agent does the work, creates a task, and notifies the person to sign it off. The judgement stays with your people. The fetching and typing does not.
The one sentence to remember: a chatbot answers questions. An agent does the work and brings you the result to approve.
Who is in control
The worry we hear most: what if the AI misbehaves, and do I have any control? Here is the honest answer, part by part.
| The question | The answer |
|---|---|
| What always waits for a person? | Every message a customer sees waits for a person. The agent does the work, creates a task, and notifies the person to sign it off. Quotes, counter-offers, tracking answers, anything a customer reads. |
| What can the person do at the sign-off? | At the sign-off, the person can edit the message, approve it, or reject it. It is an editable panel, and that is the point of it. Every version of a quote is kept, and the line that changed is shown. |
| Whose name goes on it? | An approved message goes out in the person’s own name, with their own signature, from their own email. No customer ever sees that AI works behind it. |
| What runs on its own, and under whose rules? | Autonomy is set per task type: supervised, where a person signs, or automatic. Every message a customer sees is locked to supervised. Chasers, reminders, tracking checks and quote expiry can be set to run on their own. The rules and timing for chasers and reminders are set by the head of operations or an admin. Quote expiry rules are set by the pricing executive. Chasers and reminders can be switched off for one customer or one lane. |
| Who decides who may approve what? | Role-based access control: the admin decides who can see, edit and approve what, role by role. A junior desk can be allowed to draft and not to send. |
| So what does 85% mean? | The 85% is effort taken out of the cycle: the reading, fetching, drafting and chasing. A person still signs every customer-facing message. |
| What is not there yet? | There is no pause switch yet for one customer, one lane or the whole company. Today the person at the sign-off is that control: nothing a customer sees goes out without a person. Such cases are the exception. There is no automatic margin floor yet. Today it is the staff’s job at the sign-off to keep a quote above the margin. Automatic floors are planned. |
- What always waits for a person?
- The answer
- Every message a customer sees waits for a person. The agent does the work, creates a task, and notifies the person to sign it off. Quotes, counter-offers, tracking answers, anything a customer reads.
- What can the person do at the sign-off?
- The answer
- At the sign-off, the person can edit the message, approve it, or reject it. It is an editable panel, and that is the point of it. Every version of a quote is kept, and the line that changed is shown.
- Whose name goes on it?
- The answer
- An approved message goes out in the person’s own name, with their own signature, from their own email. No customer ever sees that AI works behind it.
- What runs on its own, and under whose rules?
- The answer
- Autonomy is set per task type: supervised, where a person signs, or automatic. Every message a customer sees is locked to supervised. Chasers, reminders, tracking checks and quote expiry can be set to run on their own. The rules and timing for chasers and reminders are set by the head of operations or an admin. Quote expiry rules are set by the pricing executive. Chasers and reminders can be switched off for one customer or one lane.
- Who decides who may approve what?
- The answer
- Role-based access control: the admin decides who can see, edit and approve what, role by role. A junior desk can be allowed to draft and not to send.
- So what does 85% mean?
- The answer
- The 85% is effort taken out of the cycle: the reading, fetching, drafting and chasing. A person still signs every customer-facing message.
- What is not there yet?
- The answer
- There is no pause switch yet for one customer, one lane or the whole company. Today the person at the sign-off is that control: nothing a customer sees goes out without a person. Such cases are the exception. There is no automatic margin floor yet. Today it is the staff’s job at the sign-off to keep a quote above the margin. Automatic floors are planned.
That is the moderation layer: the agent works, a person signs, the rules are yours, and the customer only ever sees your name.
Department by department, with the number to watch
Each desk has its own number. Find your desk in this table. That number is the one to measure first.
| Department | The work by hand today | What the agent takes | The number to watch |
|---|---|---|---|
| Pricing | Read the enquiry. Ask for the missing weight. Hunt rates across cards and portals. Phone an agent. Work the margin. Write the quote. One to two hours. | The reading, the rate fetching, the requests to agents and the chasing, the draft, the margin check. | Quotation TAT. Quotes per person per day. Quotes won out of 100. Margin kept. |
| Operations and booking | Re-type the won quote into a job. Book. Confirm. Chase the carrier. Update the customer. | The job created from the quote, filled in. The chasers. The confirmations. The status checks. | Bookings per person per day. Chasers sent by hand. Re-types per job. |
| Documentation | Bills, manifests, the missing copy, the wrong weight found late. | Reading any format. Spotting what is missing. The draft asking for it. | Documents per job by hand. Errors found late. |
| Customer service and tracking | “Where is my cargo?” answered by someone opening three portals. | The answer drafted from carrier status on the customer’s own thread, for approval. | Minutes to answer. Updates sent before the customer asks. |
| Accounts | Match the carrier bill line by line. Chase payments. Re-key into accounts. | The bill read and matched to its job before booking. Payment follow-ups on their own. | Bills matched per day. Follow-ups by hand. Days to get paid. |
| Sales | Enquiries lost in the inbox. Follow-ups forgotten. | Every mailbox watched. Every thread followed to the end. Follow-ups on their own. | Enquiries answered the same day. Follow-ups missed. |
| Management | Growth means hiring. | The same team handles more. | Shipments per person per month. Revenue per person. Hires per growth step. |
- Pricing
- The work by hand today
- Read the enquiry. Ask for the missing weight. Hunt rates across cards and portals. Phone an agent. Work the margin. Write the quote. One to two hours.
- What the agent takes
- The reading, the rate fetching, the requests to agents and the chasing, the draft, the margin check.
- The number to watch
- Quotation TAT. Quotes per person per day. Quotes won out of 100. Margin kept.
- Operations and booking
- The work by hand today
- Re-type the won quote into a job. Book. Confirm. Chase the carrier. Update the customer.
- What the agent takes
- The job created from the quote, filled in. The chasers. The confirmations. The status checks.
- The number to watch
- Bookings per person per day. Chasers sent by hand. Re-types per job.
- Documentation
- The work by hand today
- Bills, manifests, the missing copy, the wrong weight found late.
- What the agent takes
- Reading any format. Spotting what is missing. The draft asking for it.
- The number to watch
- Documents per job by hand. Errors found late.
- Customer service and tracking
- The work by hand today
- “Where is my cargo?” answered by someone opening three portals.
- What the agent takes
- The answer drafted from carrier status on the customer’s own thread, for approval.
- The number to watch
- Minutes to answer. Updates sent before the customer asks.
- Accounts
- The work by hand today
- Match the carrier bill line by line. Chase payments. Re-key into accounts.
- What the agent takes
- The bill read and matched to its job before booking. Payment follow-ups on their own.
- The number to watch
- Bills matched per day. Follow-ups by hand. Days to get paid.
- Sales
- The work by hand today
- Enquiries lost in the inbox. Follow-ups forgotten.
- What the agent takes
- Every mailbox watched. Every thread followed to the end. Follow-ups on their own.
- The number to watch
- Enquiries answered the same day. Follow-ups missed.
- Management
- The work by hand today
- Growth means hiring.
- What the agent takes
- The same team handles more.
- The number to watch
- Shipments per person per month. Revenue per person. Hires per growth step.
TAT means turnaround time. Here it is the time from the enquiry landing in your inbox to the quote going out. Forwarders use the short form, so we do too.
Topline, bottomline, running cost, and how your customers feel
Four things move. We have one customer who signed off their numbers. Here they are, named, and nothing else dressed up as an average.
- Topline. The first credible quote usually wins. Faster quotes win more of them. At Infinity Logistics, average response time fell from more than 60 minutes to about 15 minutes, and the win rate rose from 35% to 73%.
- Bottomline. Fewer hands per shipment and fewer mistakes means more margin kept. At Infinity Logistics, the quotation error rate fell from 67% to under 5%. At Infinity Logistics, revenue rose 27.5% over the same period.
- Running cost. Growth without a new hire for every step. Infinity Logistics quoted 70% more shipments with the same five-person team, comparing Q4 2024 with Q4 2025.
- Your customers. Answers in minutes instead of hours. Updates before they ask. The same thread from enquiry to invoice.
One company, one year, one signed reference. Not a promise about yours. The market-wide number does not exist yet.
Seven risks, named, and what counters each one
A page that hides risk is a pitch. Here is what can go wrong with any AI system in a forwarding office, including ours, and what counters it.
| The risk | What can go wrong | What counters it | Which programme helps |
|---|---|---|---|
| Wrong answers | The tool prices a lane wrong, or promises a date it cannot keep. | A person approves every reply before it sends. Every quote version is kept. Ask any vendor to show you one it got wrong. | The session. The workshop. |
| Your data | Your rates, customers and margins end up where they should not. | Ask where the data lives, who can see it, and whether it trains anything for anyone else. In the Playground your numbers are never shown to another firm. | The assessment. The benchmark. |
| Dependence | You stop, and you cannot get your data out. | Ask for the export before you sign. Keep your TMS as the record if you want. | The checklist. The session. |
| Adoption | The team does not use it. | One task screen instead of the inbox. People approve instead of compose. The desk staff try it on their own enquiry first. | The workshop. |
| Half-done adoption | Bought, set up, never really running. | Lane by lane, with the team in the room, until it genuinely runs on your data. | The TAT reduction programme. |
| Buying wrong | An old system with AI buttons, sold as AI. | The five tells below, and the twelve questions. | The webinar. The checklist. |
| Doing nothing | Another year of quotes at one to two hours each, while the first credible quote keeps winning. | Know your number. Then decide. | The assessment. The benchmark. |
- Wrong answers
- What can go wrong
- The tool prices a lane wrong, or promises a date it cannot keep.
- What counters it
- A person approves every reply before it sends. Every quote version is kept. Ask any vendor to show you one it got wrong.
- Which programme helps
- The session. The workshop.
- Your data
- What can go wrong
- Your rates, customers and margins end up where they should not.
- What counters it
- Ask where the data lives, who can see it, and whether it trains anything for anyone else. In the Playground your numbers are never shown to another firm.
- Which programme helps
- The assessment. The benchmark.
- Dependence
- What can go wrong
- You stop, and you cannot get your data out.
- What counters it
- Ask for the export before you sign. Keep your TMS as the record if you want.
- Which programme helps
- The checklist. The session.
- Adoption
- What can go wrong
- The team does not use it.
- What counters it
- One task screen instead of the inbox. People approve instead of compose. The desk staff try it on their own enquiry first.
- Which programme helps
- The workshop.
- Half-done adoption
- What can go wrong
- Bought, set up, never really running.
- What counters it
- Lane by lane, with the team in the room, until it genuinely runs on your data.
- Which programme helps
- The TAT reduction programme.
- Buying wrong
- What can go wrong
- An old system with AI buttons, sold as AI.
- What counters it
- The five tells below, and the twelve questions.
- Which programme helps
- The webinar. The checklist.
- Doing nothing
- What can go wrong
- Another year of quotes at one to two hours each, while the first credible quote keeps winning.
- What counters it
- Know your number. Then decide.
- Which programme helps
- The assessment. The benchmark.
Five ways to tell an old system with AI buttons from a system built around the work
Most systems a forwarder will be shown are older systems with AI features added. A few are built AI-first. You do not need a list of vendors. You need five tells.
| The tell | An add-on looks like | AI-native looks like |
|---|---|---|
| Where the work happens | A chat box beside the same old screens. | In the mailboxes and one task screen. Your people work there, not in the inbox. |
| Who does the fetching | You still open the portals. The AI summarises. | The agent pulls rates, history and status itself. Nobody hops between websites. |
| Does it act, or only answer | It answers a question you typed. | It reads, drafts, chases and hands over on its own, and brings you the result. |
| Who approves | Nobody, or everybody. | A named person, beside the original thread. It goes out in their name. |
| Between desks | Forward, copy, re-type. | The work passes with everything attached. |
- Where the work happens
- An add-on looks like
- A chat box beside the same old screens.
- AI-native looks like
- In the mailboxes and one task screen. Your people work there, not in the inbox.
- Who does the fetching
- An add-on looks like
- You still open the portals. The AI summarises.
- AI-native looks like
- The agent pulls rates, history and status itself. Nobody hops between websites.
- Does it act, or only answer
- An add-on looks like
- It answers a question you typed.
- AI-native looks like
- It reads, drafts, chases and hands over on its own, and brings you the result.
- Who approves
- An add-on looks like
- Nobody, or everybody.
- AI-native looks like
- A named person, beside the original thread. It goes out in their name.
- Between desks
- An add-on looks like
- Forward, copy, re-type.
- AI-native looks like
- The work passes with everything attached.
Ask every vendor to show each tell on one real enquiry, start to finish. Ask FreighAI the same. You judge what you see.
Real companies, real work
Two kinds of example. Our one named customer, with signed numbers. And public examples from the wider industry, each with its source and date beside it.
Infinity Logistics, a UAE freight forwarder and integrator key-account reseller, runs on FreighAI. Infinity Logistics quoted 70% more shipments with the same five-person team, comparing Q4 2024 with Q4 2025. Their story is on this site, and their numbers are the ones quoted above.
Global forwarders, carriers and regional players are using AI on quoting, documents, tracking and invoices. Twelve public examples are on the Learn page for this topic, with the publisher and the date beside each one. Where a number comes from a vendor and not the company, the page says so. More customer stories are added as they are signed.
Look at one enquiry and see where AI fits in your business
Your own work is the best textbook. Five steps, a notepad, one morning. No software needed.
- 01Follow one enquiry from landing to quote to job to invoice. Every desk it touched. Every wait.
- 02Count three things. Turnaround. Touches. Hands.
- 03Split waiting from working. The work is minutes. The waiting is hours.
- 04Split fetching from judging. Fetching and typing can be done for you. Judging stays with a person.
- 05Circle the fetching. That is where AI fits in your business. That is the list to bring to any vendor.
What a year of not knowing costs
Nobody should sit on this gap for another year because nobody explained it to them.
A pricing desk of four people gets 20 enquiries a day. At one to two hours each, that is 20 to 40 hours of quoting a day. Where the rate can be fetched without waiting on a person, the same quote goes out in 10 to 15 minutes. The four people check and approve instead of typing.
That gap is there every working day. The Playground exists so that you can see it in your own numbers, and then decide for yourself.
Three programmes, one for each job
Every one is one to one and by application. No class to join, no seats, no dates to miss.
AI Impact Assessment
A structured assessment of your process, systems, roles and data. You leave with a written blueprint for growing shipments and revenue without hiring.
Quotation TAT Masterclass
The two leaks in every quotation, measured on your own desk. One of your RFQs run live, then a before-and-after report and a blueprint.
Pricing Desk Workshop
A working account built on your own rate cards before you sit down. Your desk prices real enquiries from your own inbox, and keeps the account.
When a firm wants it to stick
Not one of the three. This is what comes next, once somebody in the firm has been through one of them.
Free, open, and no email asked for
None of this is a gate in front of the programmes. It is here because some people would rather read before they put their hand up.
The course: six lessons
What an agent does, where it pays desk by desk, the risks and their counters, add-on or AI-native, what other forwarders do, and how to map your own work.
Twelve questions to ask any vendor
One sheet. What to ask, what a good answer sounds like, and the warning sign in a bad one. It works on us too.
AI in Freight, the webinar
The same course told live in forty-five minutes. Nothing to prepare, and the recording by email if you miss it.
What “a person approves” looks like
This is one screen from an AI-native system, ours. It is here to show the rule, not to sell you anything. The work is done. One of your people says yes.
- The email is read for youEvery mailbox, every thread, followed from start to finish.
- The work behind it is doneRates found, agents chased, history checked, the reply drafted.
- A person approvesOne decision, beside the original thread. It goes out in their name.
The twelve questions, for when you are in front of a vendor
Once you understand the technology, judging a vendor is the easy part. The checklist is one sheet. It works on any vendor, including us.
Twelve questions in four parts. Does it work on my business? What will it take? What could go wrong? Who am I dealing with? Each question has a longer article on what a good answer sounds like.
How these pages are written
- Every programme is one to one: one company at a time, never a room with another company. Bring whoever else from your firm belongs there.
- You apply, a person reads it, and we call you before anything is booked. You hear back either way.
- If it fits, you get your own space in FreighAI with a slice of your own work in it, and somebody walks you through it step by step.
- The session is about your work. You will see FreighAI, and at the end we say plainly what is free and what is paid. Nobody is obliged to buy.
- Everything written on this site stays open and free to read, with no email asked for.
Common questions
Is the Playground really free?
Yes. The three programmes are free and so is everything written on the site. The only exception is the three-month TAT reduction programme, whose cost is not fixed yet and is agreed in writing before anything starts.
Why do I have to apply? Why is there no button?
We are a small team and every session is run by a person, on your own work. So we read each application and talk to you first. It is not about keeping people out. It is about not wasting your morning on something that was never going to fit.
Can I bring somebody else from my company?
One to one means one company at a time, not one person. Bring anyone else from your firm who should be there, as long as this is the right programme for their job. You are never in a room with another company.
What do I actually get access to?
You get your own space in FreighAI, with a slice of your own work in it, and a person walking you through it. Not a recorded demo and not somebody else's sample data. We agree with you first exactly what is loaded, it is used for your session only, and you can ask us to delete it afterwards.
What if you say no?
Then we say it plainly and quickly, and we tell you what would help instead. Usually that is a different programme for your role, or the free course, which asks for nothing at all.
Is this a sales call with a nicer name?
The session is about your work, not our slides. You will see FreighAI, because that is the thing you are putting your hands on, and at the end we will tell you plainly what is free and what is paid. Nobody is obliged to buy anything.
I do not understand AI at all. Where do I start?
Start with the first table on this page: five things an agent does, and one rule. Then find your desk in the table further down and measure that number. Then apply for the programme that matches your job.
Is this about FreighAI?
You do get your hands on FreighAI, because that is the system we can put in front of you. But the five tells and the twelve questions on this site work on any vendor, including us, and nobody is asked to buy anything.
- The Playground narrative, an internal working paper of 7 September 2026 (doc 27). Every section of this page follows it. It is published as the explainers in the course.verified 2026-09-07
- Signed customer reference, Infinity Logistics, 15 May 2026, as published in the case study. Every Infinity number on this page is quoted from the claims register built on it.verified 2026-08-31
No price, no date and no seat count appears anywhere in the Playground, because there are none: every programme is one company at a time and by application. The cost of the three-month programme is not fixed and is agreed in writing before anything starts. Public examples from other companies carry their publisher and date beside each one in the course.
Where we stand, on the record
The policy, the government data and the technology all arrived within a few years of each other. That window is open now, and it is open to any forwarder willing to spend a morning on it.
FreighAI has been an authorised ULIP user since August 2026 and works with its APIs directly: port and container tracking, customs status, air cargo, vehicle and FASTag data, e-way bills, and GST and IEC checks.
What we read through ULIP→ULIP exists because of the National Logistics Policy 2022, which also set the country a target: India in the top 25 of the World Bank's Logistics Performance Index by 2030, and logistics costs down towards single digits as a share of GDP.
The policy, and what we do with it→AggAiLabs Software Solutions Private Limited, with its office in Pune, India, recognised by DPIIT under number DIPP232097. That is a public record, and you can check it without asking us.
The company behind FreighAI→Infinity Logistics put their own numbers in writing in May 2026 and let us publish them. Every figure on this website comes from that document, and we name them every time we use one.
Read the reference→Being an authorised ULIP user means we are licensed to read the government’s logistics data through its APIs. It does not mean the government endorses FreighAI or has any view on this Playground.
Put your hands on it, in your own job.
Apply in ten minutes. A person reads every application and comes back within 24 hours, either way. If it fits, your own work goes into your own space and somebody walks you through it.