Bheem.TradeBheem.Trade
Paper Mode
Market Closed
Connected
IST
AcademyCourseSyllabus
BHEEM ACADEMY · COURSE APA-100

Agentic Trading

A course in price action and AI trading agents. The market-structure backbone mirrors the way Al Brooks teaches it — and every lesson is split: the concept you learn, and what the agent does with it.

22 modules9 tiers109 lessons22 labs + capstone~10 weeks

Syllabus

9 tiers · 22 modules
YOU LEARNthe price-action concept, on real charts
THE AGENTwhat the agent does with that concept, live

Every lesson below is split into two — what you learn, and the agentic-trading panel showing what the agent does. Each module ends with a hands-on lab and a gate.

0
TIER 0

Orientation

Set the mental model before the first chart.

M0

Why Price Action + Why Agents

~2 hrs
  1. 0.1
    The premise of price actionEvery indicator is derived from price; price action reads order flow directly.
    AGENTWatch an agent read the tape off raw price — no indicators, just the bars.
  2. 0.2
    Why discretionary traders failThe two killers — missed setups and emotional execution.
    AGENTSee the agent remove both: it never misses a bar and never moves a stop.
  3. 0.3
    The Trader’s EquationOnly trade when reward × P(win) > risk × P(loss).
    AGENTEvery agent setup card scores the equation for you, live.
  4. 0.4
    The agent thesisYou become the strategist; the agent becomes the disciplined trader.
    AGENTMeet Kodee — the agent that watches every bar so you don’t have to.
  5. 0.5
    Platform tourThe Hub, Quick Agents, Watch Kodee Work, the Canvas.
    AGENTRun your first agent end to end and read its result.
LABLab 0 — Run the Market Pulse and Day Read agents; write a 5-line observation journal.
GATEFirst agent run logged
I
TIER I

Reading the Chart

The raw materials — bars, pullbacks, pressure and gaps.

M1

The Bar & Signal Bars

~5 hrs
  1. 1.1
    Anatomy of a barOpen, high, low, close; body vs tails.
    AGENTThe Bar Classifier labels every bar’s shape as it prints.
  2. 1.2
    Trend bars vs doji barsSignal bars and entry bars; strong vs weak.
    AGENTThe agent tags each bar trend/doji and scores its strength 0–100.
  3. 1.3
    Signal bar → entry bar → stopEnter on a stop one tick beyond the signal bar.
    AGENTThe agent marks the signal bar and pre-places the entry and protective stop.
  4. 1.4
    Inside & outside bars — ii, ioi, OOConsecutive inside/outside bars and Breakout Mode.
    AGENTThe agent flags ii/ioi/OO clusters and the Breakout-Mode state.
  5. 1.5
    Always-inAlways-in-long, always-in-short, or unclear.
    AGENTThe Day Verdict agent reports the always-in direction every bar.
LABLab 1 — Classify 50 bars by hand against the Bar Classifier; reconcile to ≥90%.
GATE90% bar-classification agreement
M2

Pullbacks & Bar Counting

~6 hrs
  1. 2.1
    What a pullback isA pause or counter-move inside a trend.
    AGENTThe agent detects when a trend pauses and a pullback begins.
  2. 2.2
    Bar counting — H1–H4 / L1–L4Counting the legs of a pullback.
    AGENTThe Pattern Firing Feed counts the legs and labels each H/L live.
  3. 2.3
    The two-legged pullbackH2 and L2 — the staple with-trend entry.
    AGENTThe Setup Check agent fires on H2/L2 with entry, stop and target drawn.
  4. 2.4
    Double-top bear flags & double-bottom bull flagsHigh-4 / Low-4 reversals inside a trend.
    AGENTThe agent flags double-flag reversals as they form.
  5. 2.5
    The endless pullbackWhen a pullback becomes a trend.
    AGENTThe agent re-checks on a lower timeframe when a pullback won’t end.
LABLab 2 — Watch the Pattern Firing Feed for a full session; journal every fire.
GATESession journal, ≥15 classified fires
M3

Pressure, Gaps & the Moving Average

~5 hrs
  1. 3.1
    Buying & selling pressureReading who is in control from the bars.
    AGENTThe agent reads consecutive-bar pressure and reports who’s winning.
  2. 3.2
    The 20-bar EMAThe one reference line.
    AGENTThe agent tracks price vs the 20 EMA and the gap between them.
  3. 3.3
    GapsGap up / gap down; gaps mean strength.
    AGENTThe agent detects opening gaps and scores them for strength.
  4. 3.4
    Moving-average gap bars & exhaustion gapsWhat a gap from the EMA signals.
    AGENTThe agent flags MA-gap and exhaustion gaps as trend-tiring signals.
  5. 3.5
    StairsHow gaps lead to trends.
    AGENTThe agent recognises the stairs pattern and projects the next leg.
LABLab 3 — Tag pressure, the EMA relationship and gaps on 10 replay charts.
GATEPressure/gap tagging on 8/10 charts
II
TIER II

Market Structure · The Market Cycle

The backbone — where is the market, and what comes next.

M4

The Market Cycle

~5 hrs
  1. 4.1
    The four-part cycleBreakout → channel → trading range, and back again.
    AGENTThe agent reports which phase the market is in, every bar.
  2. 4.2
    Always trend or trading rangeThe first decision behind every read.
    AGENTThe agent answers it continuously: trend or range.
  3. 4.3
    Binary decisionsEach question narrows the read.
    AGENTWatch Kodee Work shows the agent narrowing the read, step by step.
  4. 4.4
    The 80% ruleMost breakouts and most reversals fail.
    AGENTThe agent weights every signal by its historical failure rate.
  5. 4.5
    InertiaThe market resists change.
    AGENTThe agent assumes continuation until the structure says otherwise.
LABLab 4 — On 10 replay charts, call the cycle phase; the agent grades each.
GATE8/10 correct cycle-phase calls
M5

Trends

~6 hrs
  1. 5.1
    What a trend isBull needs higher lows; bear needs lower highs.
    AGENTThe agent confirms a trend from the higher-low / lower-high sequence.
  2. 5.2
    Trend from the openThe day that trends from the first bar.
    AGENTThe Day Verdict agent calls a trend-from-the-open in the first bars.
  3. 5.3
    Spike-and-channel trendsThe strong opening move, then the channel.
    AGENTThe agent identifies the spike, then tracks the channel that follows.
  4. 5.4
    Tight-channel & small-pullback trendsThe strongest, hardest-to-enter trends.
    AGENTThe agent grades trend strength and flags tight, hard-to-enter trends.
  5. 5.5
    Trending trading ranges & broad-channel trendsA trend built of stacked ranges.
    AGENTThe agent maps a trend made of stacked ranges.
LABLab 5 — Tag the trend type on 10 replay charts.
GATETrend-type tagging on 8/10 charts
M6

Breakouts

~6 hrs
  1. 6.1
    What a breakout isThe strongest form of a trend.
    AGENTThe agent detects the breakout bar and the level it broke.
  2. 6.2
    Follow-throughA breakout needs a second leg.
    AGENTThe agent watches for the second leg that confirms the breakout.
  3. 6.3
    Strong vs weak breakoutsThe signs that tell them apart.
    AGENTThe agent scores breakout strength from the bars and follow-through.
  4. 6.4
    The breakout test & the failed breakoutTrend resumption, or reversal?
    AGENTThe agent flags the breakout test and calls failure early.
  5. 6.5
    Traps2nd-leg traps, surprise bars, give-up bars.
    AGENTThe agent warns on 2nd-leg traps and surprise/give-up bars.
LABLab 6 — Journal 15 breakouts: strong or weak, did it follow through.
GATEBreakout journal, ≥15 classified
M7

Channels

~5 hrs
  1. 7.1
    A channel is two linesHow to draw a channel.
    AGENTThe agent draws the trendline and channel line automatically.
  2. 7.2
    The 70% ruleHow channels usually resolve.
    AGENTThe agent applies the 70% rule to call where the channel resolves.
  3. 7.3
    Channel typesTight, broad, bull and bear channels.
    AGENTThe agent classifies the channel: tight/broad, bull/bear.
  4. 7.4
    Tight channels & micro channelsThe strongest with-trend reads.
    AGENTThe agent flags micro channels as the strongest with-trend signal.
  5. 7.5
    Evolution into a trading rangeChannel breakouts usually fail.
    AGENTThe agent detects when a channel is decaying into a range.
LABLab 7 — Draw and classify channels on 10 replay charts.
GATEChannel tagging on 8/10 charts
M8

Trading Ranges

~6 hrs
  1. 8.1
    What a trading range isHallmarks — confusion and disappointment.
    AGENTThe agent identifies a range and its confusion/disappointment hallmarks.
  2. 8.2
    Buy low, sell high, scalpHow to behave inside a range.
    AGENTThe Opportunities agent surfaces fades at the range edges.
  3. 8.3
    Breakout mode (BOM)When a range is about to break.
    AGENTThe agent flags Breakout-Mode before a range breaks.
  4. 8.4
    Tight trading ranges & ledgesThe ranges to avoid.
    AGENTThe agent marks tight ranges and ledges as no-trade zones.
  5. 8.5
    The Limit-Order-MarketFading breakouts; the perverse logic of a range.
    AGENTThe agent switches to limit-order tactics inside a range.
LABLab 8 — Paper-trade a full range day with the Opportunities agent; journal each entry.
GATERange-day paper session, ≥6 trades
III
TIER III

Support, Resistance & Reversals

The levels price respects — and how trends end.

M9

Support & Resistance

~5 hrs
  1. 9.1
    S/R, not news, controls priceWhy levels move the market.
    AGENTThe agent maps every active support and resistance level.
  2. 9.2
    Types of support and resistancePrior highs/lows, the open, the EMA.
    AGENTThe agent tracks each S/R type and how price reacts to it.
  3. 9.3
    Round numbers & prior-day levelsPDH/PDL/close as magnets.
    AGENTThe agent marks round numbers and prior-day levels as magnets.
  4. 9.4
    Measured movesLeg 1 = leg 2.
    AGENTThe agent projects measured-move targets from the first leg.
  5. 9.5
    Buy-the-close / sell-the-closeReading strength into the close.
    AGENTThe agent reads closing strength and the breakout test against S/R.
LABLab 9 — Map S/R and project measured moves on 10 replay charts.
GATES/R mapping on 8/10 charts
M10

Reversals & Patterns

~6 hrs
  1. 10.1
    Minor vs Major Trend ReversalsA trendline break, then a test of the extreme.
    AGENTThe reversal detector flags MTR candidates after the break + test.
  2. 10.2
    Double tops & bottomsMost reversals are double tops or bottoms.
    AGENTThe agent detects double tops/bottoms and their flags.
  3. 10.3
    Wedges & three-pushesThree pushes into an extreme.
    AGENTThe agent counts the three pushes and draws the wedge.
  4. 10.4
    Head & shoulders, triangles, final flagsThe rest of the reversal vocabulary.
    AGENTThe agent flags these patterns as they form.
  5. 10.5
    ClimaxesBuy/sell climaxes, V tops and bottoms.
    AGENTThe agent detects climaxes and V reversals.
LABLab 10 — Identify 10 reversals on replay charts; mark the break and the test.
GATE8/10 reversals identified
IV
TIER IV

Open Interest & Options Flow

The second lens — read positioning, not just price.

M11

PCR & Max Pain

~5 hrs
  1. 11.1
    What open interest isLive contracts outstanding — positioning, not volume.
    AGENTThe agent pulls the live option chain and reports total call and put OI.
  2. 11.2
    Put-Call Ratio — reading itOI-PCR vs volume-PCR; what a rising or falling ratio means.
    AGENTThe agent computes OI-PCR and volume-PCR every refresh and trends both.
  3. 11.3
    PCR bias & extremesHigh PCR leans bullish, low leans bearish — at the edges it inverts.
    AGENTThe agent calls the PCR bias and warns when it hits a contrarian extreme.
  4. 11.4
    PCR trapsWhy a screaming PCR is often a crowded, about-to-fail position.
    AGENTThe agent flags PCR readings that contradict price as likely traps.
  5. 11.5
    Max painThe strike where most options expire worthless — and why price drifts to it.
    AGENTThe agent computes max pain and projects the expiry-day pull toward it.
LABLab 11 — Read PCR and max pain on 10 live/replay chains; journal the bias each gave.
GATEPCR + max-pain read on 8/10 chains
M12

OI Buildup & Writing Flow

~6 hrs
  1. 12.1
    The four OI statesLong buildup, short buildup, long unwinding, short covering — price × OI.
    AGENTThe agent tags each strike with its OI state from the price-and-OI change.
  2. 12.2
    Long & short buildupPrice up + OI up is fresh longs; price down + OI up is fresh shorts.
    AGENTThe agent flags fresh buildup and scores the conviction behind it.
  3. 12.3
    Long unwinding & short coveringPrice down + OI down is longs exiting; price up + OI down is shorts covering.
    AGENTThe agent detects unwinding and covering and warns when a move is hollow.
  4. 12.4
    Call writing as resistanceHeavy call OI build is sellers defending a ceiling.
    AGENTThe agent maps call-writing strikes as dynamic resistance on the chart.
  5. 12.5
    Put writing as supportHeavy put OI build is sellers defending a floor.
    AGENTThe agent maps put-writing strikes as dynamic support and tracks shifts.
LABLab 12 — Tag the four OI states across a session; mark writing-based S/R on 8 chains.
GATEBuildup tagging on 8/10 chains
M13

OI + Price-Action Confluence

~5 hrs
  1. 13.1
    OI-confirmed entriesA with-trend setup backed by supportive buildup is the higher-odds trade.
    AGENTThe agent only upgrades a setup card when OI confirms the price-action read.
  2. 13.2
    OI-confirmed exitsUnwinding or covering against your position is an early exit cue.
    AGENTThe agent flags an exit when OI starts to fight the open trade.
  3. 13.3
    When OI and price disagreeA breakout with no buildup, or a level with no writing — the warning signs.
    AGENTThe agent downgrades signals where the chain and the chart disagree.
  4. 13.4
    Writing strikes vs chart S/RWhen an option wall lines up with a prior high or low.
    AGENTThe agent highlights strikes where writing S/R and chart S/R stack up.
  5. 13.5
    The fused readOne verdict from two lenses — structure plus positioning.
    AGENTThe agent fuses the price-action and OI reads into a single Day Verdict.
LABLab 13 — For 15 setups, record the price-action read, the OI read and the fused verdict.
GATEConfluence journal, ≥15 reads
V
TIER V

The Trading Day & Execution

Putting structure to work, bar by bar.

M14

Day Types & the Open

~5 hrs
  1. 14.1
    Why day type mattersThe same setup wins on a trend day and loses on a range day.
    AGENTThe Day Verdict agent filters every setup by the day type.
  2. 14.2
    Trend, range and reversal daysReading the character of the session.
    AGENTThe agent calls the day type and updates it as the day develops.
  3. 14.3
    Trading the openOpening reversals, gap openings, the first hour.
    AGENTThe agent reads the open and the first reversal.
  4. 14.4
    The middle and end of the dayLate tests of S/R; the close.
    AGENTThe agent watches late tests of S/R and the close.
  5. 14.5
    Pre-market prepMagnets and context — how yesterday ended.
    AGENTMarket Open and Day Read agents build the pre-market brief.
LABLab 14 — For 5 mornings, call the day type at 09:45; score against the Day Verdict.
GATE≥3/5 correct day-type calls
M15

The Trader’s Equation & Trade Management

~5 hrs
  1. 15.1
    The equation in depthReward × P(win) vs risk × P(loss).
    AGENTEvery agent setup card shows entry, stop, target and R:R.
  2. 15.2
    ProbabilityPosition in the move; the 40–60 rule.
    AGENTThe agent estimates P(win) from the setup’s position in the move.
  3. 15.3
    Orders & protective stopsStop, limit and bracket orders; never widen a stop.
    AGENTThe agent places the orders and never widens a stop.
  4. 15.4
    Scaling in & outAdding to winners; the risk of scaling into losers.
    AGENTThe agent scales the position within your mandate.
  5. 15.5
    Taking profitsScalp vs swing; trailing; the exit.
    AGENTThe agent books partials and trails the runner.
LABLab 15 — Take 20 paper trades; write the equation reasoning for each.
GATE20-trade journal with equation reasoning
VI
TIER VI

The Autonomous Desk

The agents stop assisting and start doing — you supervise.

M16

The Multi-Agent Desk

~5 hrs
  1. 16.1
    Process over predictionYour edge is the routine you run every day.
    AGENTThe agent runs your routine identically every session.
  2. 16.2
    Quick Agents as a routineMarket Open → Pulse → What’s Moving → FII/DII → Day Read.
    AGENTYou chain the Quick Agents into one pre-market sequence.
  3. 16.3
    Watch Kodee WorkReading the live execution trace.
    AGENTThe agent shows every tool call and decision as it works.
  4. 16.4
    The Day Verdict as anchorOne read to orient the session.
    AGENTThe Day Verdict agent opens every session with one read.
  5. 16.5
    Designing your deskWhich agents, in what order, with what triggers.
    AGENTYou configure which agents run, and when.
LABLab 16 — Document your pre-market + intraday agent routine; run it for 3 days.
GATEDocumented routine + 3-day run log
M17

The Autonomous Trading Agent

~6 hrs
  1. 17.1
    The autonomous loopAnalysis → patterns → opportunities → execution → management.
    AGENTOne agent runs the whole loop itself, without you driving it.
  2. 17.2
    Setting the mandateSymbols, day types, setups, max risk, max trades.
    AGENTYou brief the agent; it trades only within the mandate.
  3. 17.3
    Watching it workThe live execution trace.
    AGENTThe trace shows the agent analysing and acting, in the open.
  4. 17.4
    Approve vs autoPropose-and-approve, or execute on its own.
    AGENTThe agent proposes for your tap, or executes within the mandate.
  5. 17.5
    Guardrails & the kill switchHard caps and the one-tap kill switch.
    AGENTThe agent is bound by hard risk caps you can cut instantly.
  6. 17.6
    Supervising, not drivingYou audit; you don’t place trades.
    AGENTYou audit the agent’s reasoning instead of placing trades.
LABLab 17 — Give an agent a mandate; run a full paper session in approve mode; review its trace.
GATESupervised autonomous session completed
VII
TIER VII

Build Your Agent

Stop using agents. Build one.

M18

Encode a Strategy into the Brain

~5 hrs
  1. 18.1
    From discretionary to explicitTurn a read into entry + filter + risk + exit.
    AGENTThe agent turns your words into a structured rule.
  2. 18.2
    The BrainHow the platform stores and recalls strategies.
    AGENTThe agent stores and recalls your strategy from the Brain.
  3. 18.3
    Writing a strategy in plain languageEntry, filter, risk, exit.
    AGENTThe agent reads your plain-language strategy and applies it.
  4. 18.4
    Testing recallSurfacing the strategy when conditions match.
    AGENTThe agent surfaces your strategy when the market matches it.
LABLab 18 — Write and save one personal strategy; trigger the agent to recall it.
GATEOne strategy saved & recalled
M19

Your Own Agent

~6 hrs
  1. 19.1
    Configuring a personal agentInstructions, allowed setups, risk limits.
    AGENTYou set the agent’s instructions, setups and risk limits.
  2. 19.2
    Your filtersDay types, setups, symbols.
    AGENTThe agent trades only what you allow it to trade.
  3. 19.3
    BacktestingRunning the agent over historical data.
    AGENTThe agent runs itself over history on command.
  4. 19.4
    Reading the resultsWin rate, expectancy, realised R.
    AGENTThe agent reports win rate, expectancy and realised R.
  5. 19.5
    IteratingTightening rules from the backtest.
    AGENTThe agent re-runs as you tighten its rules.
LABLab 19 — Build a working personal agent and produce a backtest report.
GATEPersonal agent deployed + backtest
VIII
TIER VIII

Monetise

Turn a working edge into a product.

M20

Fine-tune Your Model

~5 hrs
  1. 20.1
    Why fine-tuneLearning from your own outcomes, not generic data.
    AGENTThe agent learns from your own trade outcomes.
  2. 20.2
    Your journal as training dataOutcomes become the training set.
    AGENTThe agent turns your journal into a training set.
  3. 20.3
    The fine-tuning flowRunning it on the platform.
    AGENTThe agent runs the fine-tune on the platform.
  4. 20.4
    EvaluationA/B the tuned model against the base.
    AGENTThe agent A/B-tests the tuned model vs the base.
LABLab 20 — Produce a model fine-tuned on your paper-trade history; A/B vs the base agent.
GATETuned model + A/B report
M21

The Marketplace

~4 hrs
  1. 21.1
    Edge as a productWhat a buyer is actually renting.
    AGENTYour agent becomes a product other traders rent.
  2. 21.2
    PricingSubscription vs performance-linked.
    AGENTThe agent’s track record sets the listing price.
  3. 21.3
    ListingPackaging your strategy or model.
    AGENTYou publish the agent to the marketplace.
  4. 21.4
    The flywheelRenters → feedback → improvement → reputation.
    AGENTRenter feedback flows back to improve the agent.
  5. 21.5
    Compliance & honestyPaper vs live records, disclaimers.
    AGENTThe agent shows paper vs live records with disclaimers.
LABLab 21 — Publish a strategy on the marketplace (test environment).
GATEA marketplace listing published
★
CAPSTONE

The Live Desk

Run your own agent paper-live on bheem.trade for ten trading sessions. Journal every one. Present the desk you built — the agent, the strategy, the equity curve. Certification on completion; the strongest desks are featured on the marketplace.

Enrolment

Three ways to take the course.

Audit

Free
  • All video lessons
  • The full syllabus
  • No labs · no certificate
RECOMMENDED

Full + Labs

Certified
  • Everything in Audit
  • Live labs on bheem.trade
  • 22 graded gates + capstone
  • Certificate on completion

Cohort

Mentored
  • Everything in Full + Labs
  • 8-week mentored cohort
  • Live chart sessions