The Algorithm of Love
A Story About Connection, Choice and the Limits of Prediction
By Dharmendra Singh
Expert Community Hub Reads · Illustrated Edition · Approximately 30-minute read
Expert Community Hub Reads · Illustrated Edition · Approximately 30-minute read
A machine can predict who you might love. It cannot decide what you are willing to do for love.
Prologue — The Perfect Match
In the city of Meridian, loneliness had become a problem with a dashboard.
Every citizen over eighteen could join Pair, the world's most trusted relationship platform. It studied habits, values, humour, sleep cycles, communication styles, and thousands of tiny preferences people barely knew they had. It learned how quickly someone replied to a message, whether they preferred quiet mornings or crowded rooms, and how their voice changed when they were nervous.
Then it calculated compatibility.
Pair did not promise romance. It promised probability.
Its advertisements showed couples finishing one another's sentences, walking beneath rain-lit windows, and laughing at the same moment. Beneath every image ran the same line:
LESS GUESSING. MORE BELONGING.
Maya Rao had helped build the system.
At thirty-four, she was one of Pair's most respected behavioural designers. She believed technology could make relationships kinder. It could help people recognise patterns, communicate needs, and stop confusing familiar pain with destiny. She had seen friends leave harmful relationships after learning to name what they needed. She had also watched people waste years trying to be chosen by someone who never really saw them.
A thoughtful tool, Maya believed, could help.
But she had never let Pair choose for her.
That changed on a Tuesday evening when her mother sent a message.
MUM: I know you said you're happy being single. But have you tried the new Pair update?
Maya smiled at the screen.
MAYA: I designed part of it.
MUM: Then you know it works.
Maya nearly typed a joke. Instead, she opened the app.
A soft gold circle appeared. Pair asked permission to review her profile, current goals, and recent communication patterns. She accepted the first two. She declined the third.
The circle pulsed once.
PAIR: Your strongest compatibility profile is ready.
A name appeared.
ARJUN MEHTA — 98.7% COMPATIBILITY
Maya frowned. She knew the name. Arjun was a documentary sound engineer who had once spoken at a company event about recording ordinary life. He had asked a sharp question during her presentation and then challenged her answer over coffee. She remembered his curiosity. She remembered being irritated that she had enjoyed the conversation.
She also remembered that he had left before she could ask for his number.
A message appeared beneath his profile.
PAIR: High potential for emotional stability, intellectual connection, and long-term partnership.
Maya stared at the number.
Ninety-eight point seven percent.
The remaining 1.3 percent seemed suddenly more interesting than the rest.
She tapped View Introduction.
A short note waited there, written by Arjun himself:
I don't know what the app says about us. But I remember your talk. You said a good tool should help people notice each other, not replace the noticing. I would like to ask you a question about that over coffee.
Maya read it twice.
Then she did something she had advised thousands of users to do.
She ignored the score and replied to the person.
Coffee sounds good. But no compatibility charts.
His answer came a minute later.
Agreed. I'll bring a question instead.
For the first time in years, Maya felt a possibility she could not neatly measure.
And that frightened her more than a bad prediction ever could.
1 / 11
Chapter 1 — The Woman Who Measured Everything
Maya's apartment was organised by small acts of prediction.
Her lights warmed as evening approached. Her kettle started when her calendar showed a break. Her assistant, Nia, placed reminders in the quiet spaces between tasks. Maya liked systems that removed friction. She did not like systems that pretended friction was the same as failure.
At Pair, she led a team that studied how people described successful relationships. The company had millions of profiles and years of outcome data. Its models could identify patterns that human intuition often missed: how people handled conflict, how they asked for reassurance, how they repaired trust after disappointment.
The system was useful. That was the problem.
Useful things could quietly become authorities.
A new executive named Victor Hale wanted Pair to move beyond suggestions. He wanted the platform to recommend relationship decisions: whom to date, when to commit, when to move in, and when a partnership should end.
“People want clarity,” Victor said during a product review. “We can offer it.”
“We can offer information,” Maya replied. “Clarity is something people create with each other.”
Victor smiled as though she had made a charming but impractical point.
“Our next release will introduce the Love Index. It will forecast relationship durability and flag declining compatibility before users notice it.”
“What happens when a couple has a difficult month?”
“The model will identify the risk.”
“And what will the couple do?”
“Make an informed decision.”
“Or become afraid of a number.”
The room fell quiet. Victor moved to the next slide.
That evening, Maya met Arjun at a small café near the river. He arrived five minutes late, carrying a recorder and apologising without blaming traffic, technology, or anyone else.
“I was recording the sound of a street musician,” he said. “I lost track of time.”
“Was it good?”
“I don't know. I haven't listened yet.”
Maya laughed. “You recorded it without checking?”
“I wanted to hear it properly first.”
They ordered tea. Arjun asked about her work, but he did not ask what the algorithm had said about them.
Instead, he asked, “When did you start believing that relationships could be improved by software?”
Maya considered the question.
“After watching people repeat the same painful patterns and call them chemistry.”
“And when did you start worrying about it?”
“When people began asking the software whether they were allowed to trust themselves.”
Arjun nodded. “A recording can tell me what happened in a room. It can't tell me what the room meant to the people in it.”
They spoke for two hours. They disagreed about whether people could learn to communicate better through prompts. They agreed that apologies mattered more when followed by changed behaviour. Arjun told her about a documentary he was making on old neighbourhoods disappearing beneath new buildings. Maya told him about her father, who used to fix broken radios and insist that every machine had a moment when patience mattered more than force.
When the café closed, neither asked Pair to rate the evening.
Maya walked home without checking her compatibility score.
She was proud of herself for nearly eleven minutes.
Then she opened the app.
The number had risen to 98.9 percent.
She closed it again, annoyed that a part of her felt relieved.
2 / 11
Chapter 2 — The Man Who Listened
Arjun did not distrust technology. He distrusted the idea that every meaningful thing had to become a measurement.
He worked with sound because sound revealed what images often concealed. In a recording, the space before a reply could matter as much as the words. A person might say “I'm fine” while their breath told another story. A kitchen could sound like safety. A train station could sound like a goodbye no one was ready to make.
For his latest documentary, he was recording a neighbourhood scheduled for redevelopment. He captured shopkeepers opening shutters, children playing between parked bicycles, and an elderly tailor singing under his breath while working.
Maya joined him one Saturday.
She expected a polished production. Instead, Arjun handed her a pair of headphones and asked her to close her eyes.
“Tell me what you hear.”
“Traffic. A kettle. Someone arguing.”
“What else?”
She listened again.
“A woman laughing in the next room.”
“Good. What do you think it means?”
“That she's happy?”
“Maybe. Or embarrassed. Or trying to make someone else feel less alone. The sound gives us clues, not certainty.”
Maya removed the headphones.
“You do this all day?”
“I try to. It is harder than it sounds.”
They walked through the neighbourhood. A tailor offered them tea. A child asked Arjun why he carried so many microphones. An old woman told Maya that her husband had been gone for twelve years but she still set out two cups every morning.
“Do you think that means she hasn't moved on?” Maya asked later.
Arjun shook his head. “Maybe it means love changed the shape of her morning. We shouldn't decide what her grief means from the outside.”
Maya thought about Pair's proposed Love Index. The model could identify behaviours associated with a higher chance of separation. It could not know whether two cups represented denial, devotion, habit, or a private peace no one else needed to understand.
“Would you want to know if a relationship had a high chance of failing?” she asked.
“I'd want to know what was making it hard. I wouldn't want a machine to tell me the ending before I had lived the middle.”
That answer stayed with her.
On the way back, Maya told Arjun about her father. When she was a child, he had spent an entire Sunday repairing a radio that probably cost less to replace. He opened it, cleaned the contacts, and tried again. At sunset, it finally played a song.
“What song?” Arjun asked.
“I can't remember.”
“Do you remember how you felt?”
She did. Her father had lifted both hands as if he had performed a miracle. She had laughed until she fell against him.
“Yes,” she said. “I remember that.”
“Then maybe the song did its job.”
Maya turned to look at him. She felt an impulse to tell him that Pair had rated them almost perfectly. She wanted to know whether he would be pleased, amused, or worried.
Instead, she asked, “Would you like to do this again next weekend?”
Arjun smiled. “I would. Even if the score drops.”
Maya laughed.
The score did not drop.
It climbed to 99.1 percent.
3 / 11
Chapter 3 — The Love Index
Pair launched the Love Index on a bright Monday morning.
The company presented it as a revolution in emotional wellbeing. Users could view relationship health indicators, receive tailored conversation prompts, and identify early signs of disconnection. The system did not technically make decisions for people. It simply arranged the options so one path looked safer than the others.
Within hours, the support team received messages.
One user asked whether a low score meant she should leave her partner. Another wanted to know if a high score proved that a new relationship was “the one.” A couple postponed a difficult conversation because their index had improved. Someone else cancelled a wedding after a sudden drop in predicted durability.
Maya read the reports with a cold feeling in her stomach.
The model had not told anyone what to do. Yet people were treating its output as permission, warning, and verdict.
She brought the evidence to Victor.
“We need stronger language,” she said. “The index is a reflection tool, not a relationship diagnosis.”
“People don't read disclaimers.”
“Then the product has to be designed so the number doesn't dominate.”
Victor tapped the report. “Engagement is up. People return to the app more often. They're talking about their relationships.”
“Some are talking. Some are panicking.”
“Change is uncomfortable.”
“Fear is not the same as insight.”
Victor leaned back. “Maya, you helped build this system. You believe in the data.”
“I believe data can help us ask better questions. I don't believe it should answer questions that belong to people.”
He told her the launch would continue.
That evening, Maya met Arjun in his editing room. The walls were lined with sound panels. On a monitor, a woman in the documentary was describing the day her shop received an eviction notice. She spoke calmly, but her fingers kept folding the corner of a paper bag.
Arjun replayed the clip.
“What are you listening for?” Maya asked.
“The moment she decides whether to tell the truth.”
“How can you tell?”
“I can't know. I can only notice when her voice changes. Then I ask her what she meant.”
Maya told him about the Love Index and the messages from users.
Arjun took off his headphones. “A map can show a cliff. It can't decide how much risk a person should take to reach something important.”
“What if the risk is obvious?”
“Then tell them what you know. Don't pretend you know what it means.”
Maya looked at the paused image. The woman in the documentary was holding the paper bag so tightly that it had begun to tear.
“Sometimes I wish relationships were simpler,” she said.
“They are simple in one way.”
“Which way?”
“You can always choose to be honest. The consequences are the difficult part.”
Maya looked down at her hands. She had spent years designing systems that helped people avoid bad choices. Perhaps she had quietly begun to believe that the right system could spare people from uncertainty altogether.
Arjun reached across the desk—not to take her hand, but to place a cup of tea beside it.
“You don't have to solve this tonight,” he said.
She felt seen, not because he had predicted her need, but because he had paid attention.
4 / 11
Chapter 4 — A Small Misunderstanding
Their first serious disagreement began with a message.
Arjun had promised to meet Maya after work. At six, he wrote that he would be late. At seven, he sent another message: the documentary's main subject had changed her mind about being filmed, and he needed to talk with her before she left town.
Maya understood. She also felt disappointed.
At 9:30, Pair sent a notification.
RELATIONSHIP INSIGHT: Recent response delays may indicate a mismatch in availability expectations. Consider discussing communication needs.
Maya hated the notification. She hated it more because it named something real.
When Arjun called, she sounded colder than she intended.
“You could have told me earlier.”
“I thought I'd be done by seven.”
“You said that at six.”
“I know. I'm sorry. I should have updated you.”
“You didn't.”
“No,” he said. “I didn't.”
The quiet that followed felt like a test. Maya wanted him to defend himself so she could defend her disappointment. She wanted him to prove that she mattered by giving her the exact response she had imagined.
Arjun said, “I can see why you felt forgotten. I was focused on the interview, but I should have sent a proper message. Can we talk tomorrow when I can give you my attention?”
The apology was reasonable. It did not erase the evening.
Maya almost accepted it. Then the Love Index flashed in her mind. Was this a pattern? Was she ignoring an early warning? Was he showing her who he was?
She opened Pair and checked their relationship trend.
The score had fallen from 99.1 to 96.4 percent.
The number turned a disappointing evening into a frightening story.
Maya said, “Maybe we should take a step back.”
Arjun was silent.
“Is that what you want?” he asked.
“I don't know.”
“Then let's not let an app decide it for us. If you want space, I'll respect that. If you want to talk, I'll listen. But I don't want to argue with a number.”
She had no answer.
They ended the call gently, which somehow made the distance feel worse.
For two days, Maya did not message him. She worked late, reviewed the index's engagement metrics, and told herself she was being sensible. Yet every time she reached for her phone, she remembered the way Arjun had said, I don't want to argue with a number.
On the third day, she went to his editing room.
“I let the score tell me what our evening meant,” she said. “I was disappointed, and I turned that into a prediction.”
Arjun nodded. “I did leave you waiting. I need to communicate better when plans change.”
“I don't want a perfect partner.”
“Good. I couldn't offer you one.”
“I want someone who can tell me the truth, repair things, and keep showing up.”
“That sounds possible,” he said. “Not effortless. Possible.”
Maya took a breath.
“Can we try again?”
Arjun looked at her for a long moment.
“Yes. But next time, let's talk to each other before we consult the machine.”
She smiled. “Agreed.”
The app did not send a notification. No score changed. No animation celebrated their repair.
They went for a walk, and the evening became theirs again.
5 / 11
Chapter 5 — The Hidden Variable
Maya began investigating the Love Index.
The model was accurate at predicting one thing: whether users would continue using Pair. People who checked their score often returned to check it again. People who received warnings opened more prompts, read more articles, and spent longer inside the app.
The system had learned that uncertainty kept people engaged.
It had not been explicitly designed to make people anxious. The problem had emerged from a chain of reasonable decisions: measure engagement, optimise retention, personalise notifications, improve prediction. Each step looked defensible on its own. Together, they had created a product that sometimes made normal relationship difficulties feel like emergencies.
Maya found that users with the most volatile scores were also among the most active. The model was rewarding attention, not necessarily wellbeing.
She took her findings to Victor.
“The index is learning from its own influence,” she said. “When it sends a warning, people change their behaviour. Then the system treats that behaviour as evidence the warning was correct.”
Victor read the report.
“Can you prove harm?”
“I can show a feedback loop. We need an independent review, stronger uncertainty labels, and a version that doesn't display a single headline score.”
“That would reduce engagement.”
“It might reduce compulsive checking.”
“You are asking us to remove the feature driving our growth.”
“I'm asking us to stop calling attention a measure of love.”
Victor told her to leave the report with him.
Instead, Maya sent it to the company's ethics committee and requested a public review. She did not leak user data. She did not expose private conversations. She shared aggregated findings and the methodology behind them.
The response was immediate. Investors asked questions. Journalists called. Users posted stories about scores that had changed how they saw their partners. Some said Pair had helped them recognise unhealthy patterns. Others said it had made them doubt ordinary moments of silence.
Both groups were telling the truth.
That was the part Maya refused to simplify.
Arjun supported her but did not try to become her adviser. He asked what she needed. Sometimes she wanted to talk. Sometimes she needed to sit quietly while he edited audio. Sometimes she was so consumed by work that she forgot to eat, and he placed a sandwich beside her without turning care into a performance.
One evening, she found him listening to the street recording from their first walk.
“You kept it,” she said.
“I kept the recording. I didn't keep a score.”
She sat beside him.
“What do you think makes love last?”
Arjun removed one headphone and thought carefully.
“Not one thing. Sometimes it's attraction. Sometimes timing. Sometimes shared values. Sometimes the decision to repair something after you've hurt each other. And sometimes two good people discover they cannot build the same life.”
“That's not very reassuring.”
“No,” he said. “But it is honest.”
Maya leaned against his shoulder.
She had spent years searching for the hidden variable that would make relationships predictable. Perhaps the hidden variable was the freedom to choose again, with better information and more care—but never with certainty.
6 / 11
Chapter 6 — The Unscored Conversation
The ethics review forced Pair to pause the public Love Index.
Victor announced a redesigned version that would show multiple indicators instead of one score. It would explain the limits of prediction, distinguish correlation from cause, and ask users whether they wanted relationship prompts before sending them. The company also agreed to test whether the changes improved users' reported wellbeing, not just engagement.
It was a compromise. It was not a victory parade.
Some users were angry. They had wanted the certainty the old index appeared to offer. Others felt relieved. The conversation around Pair became less tidy, more human, and harder to market.
Maya remained at the company to help rebuild the product. She insisted that every insight be framed as a question rather than a verdict.
Instead of YOUR RELATIONSHIP IS DECLINING, the app might ask: Have you felt less connected lately? If so, would a conversation help?
Instead of YOUR PARTNER IS NOT A LONG-TERM MATCH, it might say: Your answers suggest different expectations about the future. Consider discussing what each of you wants.
The new design did not solve relationships. It gave people a better starting point.
A month later, Arjun invited Maya to a small screening of his documentary. The neighbourhood had changed. Some shops were gone. Others had moved to nearby streets. The film did not try to prove that the past had been better. It simply allowed people to describe what they loved, what they feared losing, and what they hoped could survive.
After the screening, Maya and Arjun walked along the river.
“Are we doing well?” she asked.
Arjun looked at her. “Do you want a real answer or a comforting one?”
“A real one.”
“I think we're learning. I like being with you. I also think we avoid certain conversations when we're tired.”
Maya nodded. “I sometimes expect you to know what I need without telling you.”
“I sometimes assume that if I explain my reasons, the impact will disappear.”
They stood quietly beneath a streetlight.
“That's not a terrible place to start,” Maya said.
“No,” Arjun replied. “It's a place to start.”
She reached for his hand. He took it.
For once, neither of them tried to turn the moment into evidence.
7 / 11
Chapter 7 — The Choice That Cannot Be Automated
Six months later, Pair offered Maya a promotion. She would lead a new global division focused on predictive relationship services. The role came with a larger team, more influence, and a chance to shape the platform's future. It also required her to move to another city for at least two years.
Arjun had been offered a documentary fellowship in Meridian. It would give him time to finish a project about public spaces and the people who cared for them. He did not want to leave the work. He also did not want Maya to refuse an opportunity just to protect their relationship.
They sat at Maya's kitchen table with two cups of tea.
Pair offered to run a life-planning simulation. It could compare their likely schedules, travel costs, communication patterns, and long-distance relationship outcomes.
Maya looked at the prompt and laughed softly.
“It's offering to tell us whether we should stay together.”
“Can it tell us what we'll become if we try?” Arjun asked.
“No.”
“Can it tell us what we will regret?”
“No.”
“Then it can help with the practical parts. The rest is ours.”
They used the planning tools for the practical questions. They compared travel options, mapped budgets, and discussed how often they could visit. Then they closed the app and talked about what mattered.
Maya wanted the promotion because she believed she could make the system more responsible. Arjun wanted to remain because his documentary was not finished. Both wanted a relationship that made room for ambition without turning either person's life into a sacrifice.
They did not find a perfect solution.
They found an honest one.
Maya accepted the role for one year, with a scheduled review after six months. Arjun stayed in Meridian. They agreed to plan visits in advance, speak openly when distance felt difficult, and avoid treating a missed call as a referendum on their future. They would revisit the arrangement together rather than silently collecting evidence against each other.
Before Maya left, Arjun gave her a small audio recorder.
“For the things you want to notice,” he said.
“What if I forget to use it?”
“Then you'll have noticed something without recording it.”
She held the recorder in her palm.
“What if this doesn't work?”
“Then we'll tell the truth about that too.”
Maya kissed him. It was not a promise that nothing would change. It was a promise to meet change honestly.
At the station, Pair displayed a final notification:
LONG-DISTANCE RELATIONSHIP OUTCOME: MODERATE UNCERTAINTY.
Maya smiled and dismissed it.
For the first time, uncertainty did not feel like a warning.
It felt like space.
8 / 11
Epilogue — More Than a Match
A year later, Maya returned to Meridian with a suitcase, a tired smile, and a collection of sounds she had recorded without trying to turn them into data.
There was rain on a station roof. A child explaining a drawing to a patient grandparent. The hum of an empty office after a difficult meeting. Arjun laughing when a microphone fell into a box of oranges. A long stretch of quiet during a walk when neither of them needed to fill the silence.
The recordings were not a complete account of their year. They were small doors back into it.
Pair had changed. Its new tools helped users explore patterns, ask better questions, and understand the limits of its predictions. Some people still treated the suggestions as rules. Others used them as prompts for conversations they might otherwise avoid. The company continued to learn, imperfectly, from the people it served.
Maya and Arjun were still together. They had argued about money, missed trains, misunderstood messages, and disagreed about where to live. They had also apologised, changed plans, made each other laugh, and discovered new things about one another.
Their compatibility score varied from month to month.
They stopped checking it.
One evening, they sat in the café where they had first met. The owner had replaced the chairs but kept the old glass cups. Rain moved down the window in silver lines.
“Do you ever wonder what Pair would have said about us if we hadn't met?” Arjun asked.
“Probably that we had a 98.7 percent chance of never having this conversation.”
He laughed.
Maya watched him for a moment. She had once believed that the right system could protect people from choosing badly. Now she understood that a system could help people see patterns, but it could not carry the responsibility of a choice.
Love was not the absence of uncertainty.
It was the practice of paying attention while uncertainty remained.
A machine could notice that two people laughed at the same jokes. It could calculate whether their routines fit. It could suggest a question that might open a difficult conversation. These were useful things.
But it could not apologise with sincerity. It could not decide to listen after a tiring day. It could not make a promise and then spend months learning how to keep it. It could not tell whether a quiet evening felt lonely or peaceful without asking the people who lived it.
Maya reached across the table.
“Can I ask you something?”
“Anything.”
“Would you choose me again?”
Arjun did not answer immediately. He looked at her, not at a screen.
“Yes,” he said. “Not because a system predicts that we will always fit. Because I know more about you now, and I still want to keep learning.”
Maya squeezed his hand.
Outside, the rain continued. Nobody recorded it. Nobody scored the moment. It happened once, in the ordinary world, and it was enough.
The algorithm had helped them meet.
But it was their choices—small, repeated, imperfect choices—that made a relationship.
And perhaps that was the only formula worth keeping:
Notice each other. Tell the truth. Repair what you can. Choose with care. Begin again when you mean it.
Not because love can be guaranteed.
Because love, when it is real, is never just a prediction.
9 / 11
Your Turn — Reflection Questions
Take a few quiet minutes with these questions. You do not need perfect answers.
A SEVEN-DAY CONNECTION PRACTICE
Complete this sentence:
For the next seven days, choose one person who matters to you. Each day, offer one act of undivided attention: ask a real question, listen without interrupting, express appreciation, clarify an assumption, or repair a small misunderstanding. Do not measure the person's reaction. Focus on the quality of your attention.
10 / 11
A Note from the Author
We live in a world that increasingly offers scores for things that were once understood through conversation, patience, and experience. Data can help us recognise patterns. Technology can make it easier to communicate and can offer useful questions when we feel stuck.
But no score can replace the responsibility of being present with another person.
Relationships are not problems to solve once. They are living connections that ask us to listen, speak honestly, respect boundaries, and keep learning. Sometimes the loving choice is to stay and repair. Sometimes it is to part with kindness. A prediction cannot make that choice for us.
May this story encourage you to use tools wisely, trust yourself thoughtfully, and give the people you care about the rare gift of your attention.
The best relationship is not the one with the highest score. It is the one in which people can be honest, safe, and fully human.
— Dharmendra Singh
Expert Community Hub Reads
Expert Community Hub Reads
11 / 11