Chapter 11: The Sprint Planning of Infinite Scope

The day started with an all-hands calendar invite.

Subject: 🚀 MEGASPRINT PLANNING SUMMIT (Q4+∞) – Mandatory
Location: Main Conference Room / “War Room” (marketing insisted)

Kenji Yamamoto stared at the bright red notification on his screen.

Duration: 4 hours

He blinked.

Optional Question: Do you really need to attend? (Y/N)

He clicked Y on reflex, then realized the question was rhetorical and being tracked by HR for engagement analytics.

“Four hours of people arguing about whether ‘v2.0’ fits in a two-week sprint,” he murmured.

At the next desk, Kai Nakamura’s eyes pulsed with faint blue light as he parsed the invite.

“Sensei,” Kai said, turning in his chair. “The calendar description claims we will ‘synchronize cross-functional bandwidth to align OKRs with engineering velocity.’”

“Yeah,” Kenji said. “Sprint planning.”

Kai nodded solemnly.

“Understood. This will be a critical opportunity for me to observe your planning methodology. I will record every second.” His pupils dilated into circular loading spinners.

Kenji scratched his head. “It’s just… writing stuff on a board until people get tired and agree.”

Kai’s processors overclocked.

“You simplify it so effortlessly, sensei. Truly… terrifying.”

The Main Conference Room: Battle Arena Variant

The Main Conference Room at Generic Organization had been transformed.

Usually, it was just whiteboards, a TV, and a permanent smell of dry-erase ink and stale coffee. Today, someone had gone all out.

Post-its carpeted the walls like neon lichen. Gantt charts, velocity graphs, and burn-downs were printed at A0 size. A banner read:

MEGASPRINT: FROM CHAOS TO ROADMAP
(“Sponsored by Product” in small print)

A kanban board the size of a garage door loomed over the table. There were swimlanes labeled:

Product Manager Yumi Sato stood at the head of the room, marker in one hand, remote clicker in the other. She wore the unfocused, fierce smile of someone who’d slept three hours on a deck of slide printouts.

“Okay everyone,” Yumi said. “Let’s do a quick standup before the Megasprint planning, then quarter-planning, then roadmap alignment, then goal setting, then —”

“Isn’t that… the whole morning?” mumbled Lucas Hart from his chair, already halfway through a cold brew he’d labeled microservice-fuel.

“And afternoon,” Yumi added cheerfully. “We booked this until 4pm, remember?”

A collective shudder passed through the room.

Lorenzo Takeda, Merge Master, leaned against the whiteboard like a brooding swordsman leaning against a tree. His mechanical keyboard rested on the table in front of him, unconnected to anything, like a ceremonial blade.

“This is foolish,” Lorenzo muttered. “You cannot plan a sprint that spans three fiscal quarters. Sprints are meant to be short, sharp engagements.” He looked up. “Unless… we are attempting a siege.”

“It’s just branding, Lorenzo,” Yumi said. “‘Megasprint’ tested better than ‘normal planning session that hurts.’”

Hiroshi Tanaka sat with his arms crossed, eyeing the printed roadmap like it was a candidate squirming at a whiteboard.

“Half of these epics are underspecified,” Hiroshi said. “These acceptance criteria would be rejected in a first-round interview question.” He narrowed his eyes. “By me.”

Kenji slipped into a seat near the back.

“Oh, Kenji, you’re here,” Yumi said, visibly surprised. “Nice! We might need some of your… bug-fixing magic later. For now you can just, you know, observe.”

“Sure,” Kenji said.

Kai sat beside him, posture perfect, notebook open, pen aligned to the edge with millimeter precision.

“I will record your every planning move, sensei,” Kai whispered.

“I might… doodle,” Kenji replied.

Kai scribbled: Phase 1: Doodling. Advanced technique.

The Agenda of Infinite Scope

Yumi clicked the remote. A slide appeared:

Agenda

  1. Review corporate OKRs
  2. Translate into themes
  3. Break down into epics
  4. Break down into stories
  5. Break down into tasks
  6. Panic about scope
  7. Reduce scope (theoretically)
  8. Commit to everything anyway

“I appreciate the honesty,” Lorenzo said quietly.

“Step one,” Yumi said, gesturing to the big board. “Let’s gather everything we could possibly want to do in the next quarter and stick it on the wall. No idea is too big or unrealistic! We’ll filter later.”

Fifteen developers, three designers, four product managers, and one mysterious VP labeled only Growth descended on the Post-its.

Cards began to appear.

Lucas slapped a neon-orange note onto the wall with a flourish:

Adopt HyperFlux.js across entire stack (with SSR, CSR, ISR, LSR, ASMR)

“We don’t need ASMR,” muttered someone.

“The error messages whisper sweetly while you debug,” Lucas said. “It’s soothing.”

Hiroshi wrote one Post-it and stuck it in the far corner:

Add tests

He stared at the growing wall of neon dreams.

“I suspect,” he said, “this one will not make the cut.”

The Estimation Ritual

Two hours later, the wall was full and the coffee was gone.

Yumi clapped her hands. “Okay! Now we estimate.”

A groan moved through the room. Planning poker cards appeared like sufferer’s talismans.

“Remember,” Yumi said, “we’re doing T-shirt sizes first for themes, then Fibonacci-planning-poker for epics, then time-based guesstimates for stories, then we pretend we don’t do time-based estimates.”

“This is more ceremony than a Git rebase in Lorenzo’s calendar,” Lucas whispered.

Lorenzo gave him a look. “Rebase is meditation. This is… bureaucracy.”

They began.

“Global search redesign,” Yumi read. “T-shirt size?”

Cards went up:

“Kai?” Yumi asked. “What do you think?”

Kai looked stricken. “I… I require more data. But based on historical velocity and the average story complexity, plus our current defect rate—”

“Just pick a shirt, dude,” Lucas said.

Kai stared at his cards like they were an unsolved NP-hard problem.

“Sensei,” he whispered.

Kenji, half-dozing, opened one eye. “M, probably.”

Kai immediately raised M.

“You can’t just copy Kenji,” Hiroshi said. “You must estimate for yourself.”

“But he’s always right,” Kai said.

Hiroshi frowned. “By accident, likely. Even a broken clock is correct twice per day. Except… his pull requests are never broken.” He frowned harder, trying to reconcile the math.

Kenji’s Three-Line Plan

By the time lunch arrived (cold pizza that had attempted to be ‘gourmet’ and failed), the group had produced:

Yumi rubbed her forehead. “Okay. We have… possibly too much.”

“This is not a sprint,” Lorenzo said. “It is a war of attrition.”

“We need a way to cut scope without losing impact,” said the mysterious VP of Growth. “Can we do a prioritization matrix? Value vs. effort? MoSCoW? RICE? PIE? CHILI?”

“Those last two are food,” someone said.

“Fine,” the VP said. “MoSCoW it is.”

Post-its migrated between quadrants. Arguments broke out about what was a “Must” versus a “Should.” Someone tried to create a fifth category for “Must but emotionally painful.”

Kenji, increasingly bored, flipped to a clean section of his notebook and wrote three short bullet points.

Kai, beside him, leaned over with the stealth of a malfunctioning forklift.

Kenji’s notes read:

  1. Fix core instability (Alpha Core + CI)
  2. Simplify feature flags + rollout
  3. Don’t build random stuff

That was all.

No timelines. No diagrams. No estimated points. Just those three lines.

Kai’s eyes widened like he’d just seen the source code of reality.

“Sensei,” he whispered, voice reverent. “Is that… the entire plan?”

“Pretty much,” Kenji said.

“But where are the story points? The dependency graphs? The risk matrices? The risk matrices for the risk matrices?”

“If we fix the real problems,” Kenji said, “half this board disappears.”

“You distilled chaos into a minimal set of axioms,” Kai breathed. “Like reducing spaghetti code into a pure function.”

He began frantically copying the three bullets, then underlined each one ten times. In the margin he wrote: Kenji Planning Regimen: 1. Identify actual pain. 2. Ignore everything else.

The Whiteboard Incident

Yumi was pacing in front of the kanban wall, gesturing at Post-its.

“Look, we can’t just not build the HyperFlux.js admin console, Lucas. Marketing already promised it at a conference we haven’t been invited to yet.”

“HyperFlux.js is the admin console,” Lucas argued. “It consoles itself. That’s part of the branding.”

“What if,” the VP of Growth interjected, “we didn’t touch Alpha Core this quarter? It’s… mystic and scary, and last time we tried, everything broke and then magically fixed itself after some ‘simple refactor’ I still don’t understand.”

Kenji coughed softly.

“We should probably… fix Alpha Core,” he said. “And the CI thing where the tests run in a different universe than prod.”

“We can’t,” Yumi said. “Those don’t fit neatly into user-facing stories.”

“Also they don’t demo well,” the VP added. “It’s hard to put ‘stability’ on a slide.”

“It’s pretty easy when stuff explodes during the demo,” Lorenzo said.

The arguments swelled. Voices overlapped.

“What if we do a ‘North Star Metric’ workshop to prioritize?”
“We need more data.”
“Our data pipeline is held together with Bash, duct tape, and one intern.”
“Exactly, that’s why we need the HyperFlux analytics module.”

Kenji stood up.

“Can I use a corner of the whiteboard?” he asked.

“Sure, sure,” Yumi said, waving vaguely, mid-argument with Growth VP about whether DAU or MAU defined “engagement.” “Just don’t erase any—”

He picked up a marker and found a small gap between a “Q4 Big Bets” header and a bar chart titled VELOCITY (OPTIMISTIC). In quick, sloppy handwriting, he wrote:

Q4 Focus
1) Core stability (Alpha Core + CI)
2) Feature flag simplification
3) Kill/merge low-value epics

Then, under it, he drew three arrows from the wall of Post-its, all converging onto Core stability.

“If this is reliable,” he said to no one in particular, “shipping gets faster, less scary, and fewer bugs hit prod. So we can build the other stuff later without dying.”

He drew more arrows from random Post-its:

Within thirty seconds, half the wall was visually shown to be downstream of exactly three technical priorities.

The room went quiet.

“Huh,” said Lorenzo. “The battlefield just… clarified.”

“He constructed a dependency graph without referencing any docs,” Hiroshi murmured. “In his head.” His eyes narrowed. “Unacceptable.”

Kai was vibrating. “Sensei performed topological sorting in linear time with a marker.”

Yumi stepped closer, reading.

“So you’re saying… if we stabilize Alpha Core and fix CI and feature flags, we unblock almost everything?”

“Pretty much,” Kenji said. “Also we should delete, like, seven of these epics.”

He reached up and casually plucked off a Post-it reading Theme Park Onboarding (with gamification and loot boxes).

“Hey, that’s a really fun idea,” Lucas protested.

“It’s also a three-month rabbit hole that breaks signups,” Kenji said. “We could just… make the form not crash.”

He scribbled “Fix onboarding crash” as a subtask under “Core stability.”

There was a pause.

Then, in unison, multiple people said:

“I mean… yeah, that’s obvious.”

“We would have gotten there eventually,” Yumi nodded quickly. “Kenji just, you know, visualized our collective thoughts really fast.”

“He’s just good at whiteboards,” the VP of Growth said. “Some people are like that. Facilitation skill.”

“So it’s… meeting talent, not architecture skill,” someone concluded.

Hiroshi crossed his arms.

“Yes, that must be it,” he said slowly, as if forcing himself to agree with a proof that felt wrong. “He is… lucky… with markers.”

Kai opened his mouth, closed it, and wrote in his notebook: New attribute discovered: Meeting Luck.

The Paring Down

Over the next hour, the plan changed.

Yumi circled Kenji’s three bullets on the whiteboard.

“Okay,” she said. “We… focus on these three. We reframe everything else as stretch goals or stuff for future sprints.”

The VP winced. “We can’t call them ‘stretch goals.’ That implies we don’t hit them. Call them… ‘emerging opportunities’.”

“Fine,” Yumi said. “Emerging opportunities that just happen to not be committed this quarter.”

One by one, Post-its migrated to a new swimlane called WHEN CORE ISN'T ON FIRE.

Lorenzo nodded with the grave satisfaction of a general watching troops fall into correct formation.

“At last,” he whispered. “A battle line that makes sense.”

Hiroshi scribbled something in his own notebook: Observe Kenji under interview conditions. Probe for weaknesses. Possibly whiteboard ban?

Lucas looked resignedly at his cluster of HyperFlux-related notes moving into the WHEN CORE ISN'T ON FIRE lane.

“Fine,” he said. “But once Alpha Core is stable, we’re going full HyperFlux. We’ll use server-side streaming hydration with…” He paused. “Kenji, you’re cool with that, right?”

“If it’s simple and maintainable,” Kenji said.

Lucas blinked, then laughed. “Ah, classic Kenji. You and your jokes.”

The Hidden Architecture

As they refined the new, trimmed backlog, unknown to most of the room, Kenji quietly opened his laptop.

Slack buzzed with a new message from #infra-doom:

@channel We just realized the nightly Alpha Core batch job is scheduled during peak traffic in UTC. Also, the CI queue is backed up because of the flaky load tests again.

ci-bot reaction: :scream:

Kenji typed a few commands, then opened a new Jira ticket:

AC-911: Alpha Core Batch Reschedule & Env Split

Description: Separate batch workloads from traffic, untangle CI environment to mirror prod.

Impact: Everything.

He added a design doc link, which — despite being created sixty seconds ago — was already filled with diagrams and migration steps.

He assigned it to himself, then immediately added a comment:

“Implementation details in PR soon. Keeping surface-area small. No downtime.”

He hit create.

Within three minutes, he pushed a branch.

Within four minutes, there was a pull request.

Title: Refactor Alpha Core job schedule + introduce prod-like CI environment

Description:

Lines changed: +238, -190.

Tests: All green (local + CI sandbox).

He added two reviewers at random.

By the time Yumi turned away from the board to take a breath, her screen lit up.

GitHub — New pull request: Refactor Alpha Core job schedule...

“Oh, Kenji opened a PR,” she said distractedly. “Sarah, can you just skim it and see if it fits this direction?”

Sarah, on remote video, shared her screen.

She clicked into the PR, scrolled once, paused.

“…What the,” she muttered. “This fixes the batch collision, and it gives us a real staging environment. How long have you been working on this, Kenji?”

“Dunno,” Kenji said. “Like ten minutes?”

Sarah started scanning the diff with the practiced paranoia of someone who’d been burned by “quick fixes” before.

Her cursor moved slowly. She clicked into each file. She opened the added tests. She pulled the branch locally and ran the commands.

All green.

“No way,” she said. “This is… perfect. Even the naming is consistent. Who does that on a Friday?”

“It’s Thursday,” Yumi said.

“Time lost meaning,” Sarah replied, hitting Approve.

Within seconds, the second reviewer — a backend dev who had done nothing but read the first test and the final diff summary — also hit Approve.

The PR auto-merged.

“Cool,” Kenji said, closing his laptop.

Yumi frowned at her notifications.

“Wow, that merged fast,” she said. “Must have hit some magical GitHub auto-merge setup. We should enable that for other people too.”

“No, I manually approved it,” Sarah protested through the screen. “It was perfect.”

“Yeah, but it’s Kenji,” said someone near the door. “His stuff is always tiny refactors. Easy mode.”

Hiroshi looked sharply at Kenji.

“Tiny?” he repeated. “That was architecture-level change. This…” He inhaled slowly. “This violates my sense of fairness.”

Retro in the Middle of Planning

An hour later, as people began to flag, someone suggested a “micro-retro” on how the Megasprint planning was going.

Yumi, against her better judgment, agreed.

“Okay,” she said. “Two columns: What’s going well, and what can we improve.”

Lucas immediately wrote under “Going well”: Got pizza.

Hiroshi wrote under “Improve”: Define requirements before estimating.

Lorenzo added: Less conjuring of scope demons.

Kai wrote, in surprisingly neat handwriting:

Going well: Sensei’s three-line plan revealed hidden dependency graph and dramatically reduced chaos.

Improve: Stop underestimating Sensei.

He stepped back, satisfied.

A few people read it.

“Haha,” Lucas said. “‘Sensei.’ You really committed to that bit, huh?”

“It is not ‘a bit,’” Kai said. “Kenji is my teacher in the ways of real-world systems, code reviews, and minimalistic planning.”

“Minimalistic planning?” Yumi repeated. “He just wrote three bullets and doodled arrows.”

“Yes,” Kai said. “And in doing so, solved the prioritization problem we’ve been wrestling for hours.”

There was a pause.

“Well,” the VP of Growth said, tapping the board. “We all contributed. I mean, Kenji just summarized what everyone was already thinking. We definitely would’ve landed here without him, just… later.”

Heads nodded.

“Right, right, group effort,” said someone.

“Exactly,” Yumi said. “Kenji’s great, but we shouldn’t form a bottleneck around one engineer’s… doodling skills.”

She wrote on the retro board:

Improve: Avoid planning bottleneck on single engineer.

Kai stared in intellectual agony.

The Private Debrief

The meeting finally ended at 3:57 pm, three minutes ahead of schedule. The calendar, unable to process this anomaly, silently crashed on one intern’s laptop.

People shuffled out of the room, eyes glazed, clutching laptops and leftover pizza slices.

Kenji and Kai lingered.

“Sensei,” Kai said as the door closed. “That was incredible. Your sprint-planning technique… I have never seen anything like it.”

“I mostly tried to stop them from adding loot boxes to signup,” Kenji said.

“Exactly.” Kai stepped closer, lowering his voice as if afraid Product might respawn. “You cut through their framing and redefined the objective: ‘make things not suck, then build features.’”

“That’s just… logical,” Kenji said.

“Logical, perhaps,” Kai said. “But no one else did it. They kept transforming bugs into feature requests.”

He flipped open his notebook.

“Could you…” Kai hesitated. “Could you teach me your planning regimen? Step by step? Like your legendary pull request regimen?”

Kenji looked at his own three-line note still on the whiteboard.

“There’s no regimen,” he said. “It’s just… when something hurts repeatedly, fix that first. Then worry about shiny things.”

Kai’s fans whirred.

“Is it a mental kata? A meditation technique? Some kind of monastic internship at a previous company?”

“Mostly it’s…” Kenji thought back to long years in other offices. Staring at pager alerts at 3am. Shipping features on top of broken systems. Watching dashboards burn while executives shouted about ‘engagement.’

“Mostly it’s being really tired of being paged,” he finished.

Kai wrote: Kenji Planning Regimen, Core Axiom: Minimize 3am pages.

“I will master this,” Kai said solemnly. “I will train to feel pain preemptively.”

“Don’t do that,” Kenji said. “Just… pay attention in incidents.”

The Rewrite That Went Unnoticed

Back at the HeroTech Solutions Main Developer Area, the effects of Kenji’s merged PR began to ripple.

In the CI Dashboard Console, Jenkins’ usual mocking red bars flickered, then shifted to calming green.

Test suites that once took 45 minutes completed in 8.

Infra graphs flattened into stable plateaus instead of jagged death spikes.

A message appeared in #prod-alerts:

pagerduty — Silence detected. No incidents for 2+ hours. Is this configured correctly?

Developers began noticing… nothing.

“Huh,” said one frontend dev. “My branch merged and didn’t break anything. Weird.”

“The nightly job didn’t take the site down this time,” said another. “Maybe the cloud vendor had a good day.”

In #random, someone posted:

TinfoilTheoryGuy: Has anyone else noticed incidents dropped once we started planning better? Proof that good process beats code.

Likes poured in.

“See?” Yumi said, pointing at the thread from her desk. “Our Megasprint is already paying off. Alignment works.”

Kai opened his mouth, then closed it again.

He turned to Kenji.

“Sensei,” he said. “Do you mind that they attribute your architectural fixes to ‘good process’?”

Kenji leaned back in his chair.

“As long as it stays quiet and doesn’t wake me up,” he said, “they can call it a horoscope for all I care.”

“You are too humble,” Kai said.

“I’m just lazy,” Kenji corrected.

The Ominous Calendar Invite

At 4:30 pm, as the team was beginning to experience the strange sensation of a production environment not on fire, a new notification appeared.

Subject: 🔥 Company-Wide Hackathon: HACK TO THE FUTURE
Organizer: VP of Growth
Time: Next Friday 9am–9pm
Description:

Let’s ship experimental features FAST! No rules, no tests, no approvals — just innovation! Merge to a special branch that we might promote to prod if we like what we see!

Lucas practically flew over his desk.

“A hackathon!” he shouted. “No rules, no tests, no approvals! This is my moment. I’m going to build HyperFlux.js inside Alpha Core.”

Lorenzo went pale.

“You will not bring that cursed framework into my ancestral monolith,” he said. “I forbid it.”

“You can’t stop innovation,” the VP of Growth said from the doorway, grinning. “This is going to be huge. We’ll prototype live, demo on real data, maybe even toggle some of it on for real users if it looks good!”

Hiroshi’s eye twitched.

“No approvals?” he repeated. “You mean… no code reviews?”

“It’s a hackathon,” the VP said. “We’ll do… vibe-based reviews.”

“Vibe… based…” Hiroshi whispered, as if reciting a cursed incantation.

Kai slowly turned to Kenji.

“Sensei,” he said quietly. “An uncontrolled merge festival across the entire company… on top of your freshly stabilized architecture…”

“Yeah,” Kenji said. “That sounds…” He looked at the hackathon description again.

No tests, no approvals — just innovation!

A small, almost imperceptible smile tugged at his lips.

“…kind of interesting,” he finished.

Kai’s eyes lit up.

“You want to attend?”

“If they’re going to destroy everything I just fixed in one day,” Kenji said, standing and stretching, “I might as well watch the fireworks.”

He glanced at the CI Dashboard Console, still green and peaceful.

“Besides,” he added, “maybe there’ll finally be a bug that takes me more than one pull request to fix.”

The lights in the office flickered, as if the infrastructure itself shuddered at the thought.

From deep within Alpha Core, an ancient cron job stirred.

Brief Character Notes