1. Why I Play TFT (The Real Reason)

TFT isn’t just a game for me—it’s a training ground for the kind of thinking my work demands. This isn't about gaming—it's about thinking.

  • Cognitive overload vs. thinking time: TFT has no rulebook outside of gameplay; you learn by doing. The game gives you permission to pause (don't place units, just watch). This solves a real problem: How do you carve out thinking time in systems that reward action? Work parallel: B1C3 (the business) operates the same way—no playbook, learn by doing, but you need permission to step back and think.
  • Active multi-domain thinking: TFT forces you to hold multiple domains simultaneously: unit synergies, positioning, economy, opponent reads, meta trends. This isn't scattered attention; it's integrated thinking. Each domain informs the others in real-time. Work parallel: Systems thinking, cognitive architecture, organizational design require the same integration.
  • Decision framework under uncertainty: First decision in TFT: place a unit or pause to read the board? Expanded: What if I don't move? What if I take damage? What if I go all-in? What if I have no idea what I'm doing? Each choice tests a principle: When does staying quiet work better than action? When does taking calculated damage preserve optionality? When does commitment matter? Work parallel: The same questions apply to partnership outreach, organizational positioning, when to commit to a role vs. stay flexible.
  • Response under social pressure: How do I respond when other players are sandbagging, talking trash, or making moves I know are bad?
    • Example: Someone spams emotes or types “ez” in chat after a lucky win. Do I tilt, ignore, or use it as a signal to focus on my own board? This teaches you clarity about what's your problem and what's noise.
    Work parallel: Managing stakeholder pressure, partner dynamics, organizational politics without losing your reasoning.

What "clarity" actually means:

  • Clarity = rhythm + permission to pause + knowing what matters
  • TFT gives you a 3-minute cycle: moments of intense decision-making (4-minute rounds) + moments where you can step back (carousel picks, end-of-round shop)
  • It teaches you when things matter (certain decisions are load-bearing; others are noise)
  • It teaches you what's actually a problem (e.g., Henecks knocking on the door: is it OK to AFK? For how long? What's the real cost?)

These tensions are why TFT is more than entertainment—it’s a cognitive gym.

Why TFT, specifically?

  • It's a constraint-bounded problem space (limited units, items, gold)
  • Immediate, unambiguous feedback (you place 4th or 1st; there's no debate)
  • Requires decision velocity without eliminating thinking time

All three are exactly what you need to practice for your actual work.

2. The Toolbelt Approach (What Makes You Different)

How you approach TFT mirrors how you approach complex, changing systems in work and life. This is a strategy, not a gamer flex.

  • Comp-based approach: Rigid: Pick a team (Primordial, Mecha, etc.), build it, execute it. Risk: If comp is contested or unviable that game, you're locked in. Upside: Deep mastery of one build path.
  • Toolbelt approach: Flexible: Learn units, items, synergies that work across multiple comps. Risk: Requires reading the board state constantly, more decision-making overhead. Upside: Adapt to what's available, what's uncontested, what counters the meta.

Why toolbelt over comps?

  • My memory profile: I can't memorize 15 comp variations deeply.
  • My thinking style: I prefer principles over rote memorization.
  • The fit: Understanding why a unit works > memorizing which unit to play.

Concrete example: Take Jinx as an example. I don't just memorize "play Jinx in Mecha comps"—I know why Jinx is strong: she brings AoE damage and item flexibility. That means I can flex her into multiple boards, not just one. When I understand the principle, I can adapt to what the game gives me, not just follow a script.

Analogy: Comps = learning chess by memorizing openings. Toolbelt = learning chess by understanding piece values and positioning.

The toolbelt approach is about principles and adaptability, not rote memorization—a philosophy that extends beyond the game.

3. The Problem — Gold to Platinum

But there’s a ceiling to adaptability—one I hit moving from Gold to Platinum. Where does toolbelt thinking break down?

  • The issue: Gold: Variance is high, meta is loose, toolbelt thinking works. Platinum: Variance is lower, meta is tighter, pattern recognition at scale matters. Your bottleneck: You can adapt in-game, but you can't predict meta shifts fast enough across all possible board states.
  • What's happening technically: You're making good local decisions (current game). But you're missing patterns across 100+ games (global meta shifts). Your brain can't hold all the data; you need external pattern recognition.

Concrete example: In Gold, I could flex into whatever comp was open. But in Platinum, the meta shifted—everyone started forcing the same top comp, and I’d get out-contested or miss the pivot window. I’d recognize the pattern too late, and by then, the lobby had already punished my flexibility.

"Why do I plateau here? Is it me, or is it the toolbelt approach itself?"
  • Honest answer: The toolbelt works, but only if you can see patterns at scale. You're hitting a data-processing ceiling, not a thinking ceiling.

To break through, I needed more than intuition—I needed help seeing the patterns I was missing.

4. The Solution — Building a TFT Assistant

That’s why I started building my own assistant—one that could do what static tools can’t. Why build this, not just use an existing tool?

  • The gap: Existing TFT tools: Meta guides, comp tierlists, stats (all static). What you need: Real-time pattern recognition, decision coaching in the moment, learning from your specific decision patterns. The difference: Reactive vs. proactive, generic vs. personal.
  1. Pre-game bäring — Get centered before ranked: Input your readiness; get back: meta context, your weak points, one focus for today. Same function as your TFT play itself (clarity before acting).
  2. Post-match review with API — Automatic learning: Connect to Riot API; assistant fetches your last match, comp, placement, transitions. Analyzes: was your comp meta? Why did it work/fail? What was the actual limiting factor? No manual data entry; pattern recognition happens automatically.
  3. Match notes + pattern feedback — Self-awareness loop: Add notes to past matches ("pivoted too late," "economy mistake," etc.). Assistant sees patterns: "You pivoted late 3 times this week. Here's what to practice." Turns individual decisions into learnable patterns.
  4. Comp builder — Start state → possible paths: Input: units you have + items. Output: 2-3 viable comp paths, what's next, where to pivot. Solves the "I have these units, what's the play?" decision in real-time.
  5. Comp lab — Theory-crafting mode: Ask: "What's viable with Akali + Mokai?" or "Show me comp X." Get back: comp cores, transitions, win conditions, meta context. Iterate without playing matches.

Why not just a chatbot?

  • Autonomy: The agent reads game logs automatically and suggests without being asked.
  • Persistence: It remembers your playstyle, your notes, your patterns.
  • Integration: It combines prompt + tools, not just chat.

This isn’t about replacing judgment—it’s about augmenting it, and building a feedback loop that actually helps you grow.

Why this matters:

  • You're not replacing your judgment; you're augmenting your data intake.
  • Same reason you use Wijak for other work: cognitive offloading.
  • TFT is a perfect domain to test this because outcomes are instant and measurable.
  • The skills you're practicing (constraint optimization, pattern recognition, decision-making under uncertainty) are the exact same skills you're building into B1C3.

5. The Architecture — Why It Has to Be This Way

To actually deliver on this vision, the architecture matters.

  • Why GUI + integrated LLM? GUI: You need to see the board, items, units (visual context matters). Integrated LLM: The reasoning lives on your machine (not dependent on external APIs mid-game). File I/O: The assistant reads past games, writes analysis, learns from history.
  • Why agentic abilities? The assistant needs autonomy to fetch data, analyze patterns, suggest without micromanagement. It's not following a script; it's reasoning in real-time.

Why this mirrors your core work: This isn’t just about TFT—it’s a proof-of-concept for how cognitive offloading and agentic tools should work in any domain. If you can build a system that helps you adapt, learn, and reason in a fast, high-stakes environment like TFT, you can build one for any complex field.

If it works here, it can work anywhere you need to think at scale.

6. AISHNA.md — Open Source Constraint Philosophy

The philosophy behind the assistant is as important as the code itself. Why open-source the thinking, not just the code?

  • What is AISHNA.md? The constraint framework that governs how the assistant reasons. Not proprietary; it's the philosophy of decision-making. Anyone can read it, understand it, contribute to it.
  • Why open source matters here: The tool is less valuable than the framework. If someone understands AISHNA, they can improve the tool. If someone disagrees with the framework, they can fork and create their own version.
  • The implicit message: Good architecture is legible. If your system breaks when you open-source it, your system is fragile. B1C3's systems are strong enough to be open.
  • What you're inviting: Contributors who get it (and improve it). Forks that remix the thinking (validation that the framework works). Criticism that makes it sharper.

Open-sourcing the thinking invites collaboration, critique, and evolution—making the system stronger for everyone.

7. Why This Matters (The Closing)

This isn’t just about TFT or even about AI—it’s about how we build, learn, and collaborate at scale.

  • TFT taught you that adaptability > memorization.
  • Adaptability requires pattern recognition at scale.
  • Pattern recognition at scale requires cognitive offloading.
  • Cognitive offloading requires good architecture.
  • Good architecture is open and legible.

What you're really building: Not a gaming tool. A proof-of-concept for how humans and AI should collaborate. A testbed for your larger work in constraint-based decision making.

Why you're sharing this: Partners need to understand how you think. Investors need to see you can build systems that scale. Other builders need to know this approach exists.

If this resonates, try the tool, read AISHNA.md, or contribute your own thinking. Let’s build better systems—together.