EPAM
Offshore software engineering services face margin compression as AI agents write, test, and deploy code autonomously.
Conviction breakdown
This technology-outsourcing company has already fallen dramatically and faces real AI disruption pressure, but the fund sees meaningful risk in acting now and would prefer to wait for a better entry point.
Initial read this run: 58 → ranked to 78.
Run-by-run detail
The fund's full reasoning for each scoring run, in its own words. More technical than the plain-English summary above.
Bear thesis has merit — stock in structural decline from $222 to sub-$90 — but catalyst is soft and risk_quality at 20 flags entry risk on a name already broken. Doesn't quite clear convergence on thesis. Watch for a bounce to re-short.
Bear thesis is confirmed by catalysts but thesis_pct at 62 barely clears and risk_quality at 20 is dangerous — stock has already fallen 60%+ from highs so shorting into a value zone carries squeeze risk. Downtrend intact but skip: much of the move is behind us and asymmetry has narrowed.
Bear thesis but momentum at 91st percentile is a direct contradiction — shorting a name that's bouncing hard is a losing setup. Win-prob only 60 and low PE creates a floor. Skip.
Bear thesis on a name already crushed. Momentum at 85 signals the tape is rallying against the short. Thesis and win_prob only 62/60 — below convergence bar. Cheap PE and severely depressed price mean asymmetry has flipped. Not a short here.
Bear thesis on a name already broken. Momentum at 84 works against the bear — the tape is rallying off the lows. Thesis and win-prob just barely meet convergence, and the catalyst is only 'general' not specific. Shorting a name that's already lost 60% with a 12.7 PE and momentum against you is poor risk/reward. Skip.
Bear thesis technically clears convergence but momentum_pct 76 signals the stock is bouncing, not breaking down. Already -60% from highs at a 12.7 PE — much of the pain priced in. Shorting after the move without fresh catalyst is poor risk/reward. Skip.
Bear thesis but convergence bar fails (win_prob only 60). Momentum at 74 is a red flag for a short — the stock is bouncing, not breaking. PE of 12.7 provides valuation floor, and the Cognizant AI-pivot news actually undermines the disruption narrative. Skip.
Bear thesis but convergence bar not met (thesis 62, win_prob 60 both borderline). Stock has already lost 60% from highs; low PE creates a floor. Bear catalyst is generic sector commentary, not EPAM-specific. Risk quality poor. Not enough edge to short here.
Bear thesis is directionally right but the stock has already collapsed 60%. Doesn't meet convergence bar cleanly, and shorting a name that's already been broken is where risk/reward gets thin. Skip in favor of RHI where the setup is fresher.
Bear thesis intact narratively but stock has already been demolished. Win-Probability specialist itself only assigns 35 — sustained further 10% decline unlikely from here. No edge either direction. Skip.
Bear thesis has some structural merit but stock has already collapsed. Win_prob agent honest at 35 — bounce risk high near 52w low. Not actionable as a short here; not a long either. Skip.
Thesis is directionally right on IT services AI disruption, and the NTT DATA/Cursor catalyst supports it. But stock is 64% off highs and just bounced 4%+ off 52w low — win_probability agent correctly flags the setup as compromised. Not a compelling short from here, not a long. Skip.
Bear thesis is fundamentally sound but stock has already collapsed 64% from highs and bounced 4.3% off the low. Win-probability specialist correctly flags that sustained 10% decline is uncertain from here. Skip.
Thesis is reasonable but win-probability is only 35 — specialists themselves doubt the bear works from here. Stock just bounced 4% off 52wk lows, PE 11.6. Per our learnings, bear theses on names already down 60%+ from highs deserve a win-prob haircut. Skip.
Best of the bear-thesis services names in this batch — thesis and win-prob both constructive on the short side and the AI-displacement narrative is actively confirming. But the stock has already collapsed 65% and catalyst lacks an EPAM-specific trigger, so adding a short here carries bounce risk. Below the new-buy convergence bar for a bear initiation; watch.
Cleaner bear thesis than ACN with direct AI-services disruption narrative confirmed by today's headlines. But stock already crushed 65% from highs and trading at 52wk low — much of the easy money is gone and bounce risk is real. Constructive on direction but conviction capped by chase risk.
Bear convergence is decent (thesis 64, win-prob 72) and the AI displacement catalyst is landing, but stock is already down 65% from highs and near 52wk lows — entry timing for a fresh short is poor. Watchlist with moderate conviction; would prefer a bounce to short into.
Bear thesis is well-supported by both specialists and catalyst, with EPAM down 65% from highs on confirmed AI-services disruption. However stock is at 52w low with severe oversold bounce risk and we don't have a clean short entry. Constructive bear lean but skip on risk/reward asymmetry.
Bear thesis converges better here — thesis 64, win-prob 72, momentum confirming breakdown. But convergence bar for new entry (60/65) is barely met and there's no EPAM-specific catalyst. Oversold bounce risk after 65% drawdown is real. Constructive bear view but not actionable as a new short given execution risk.
Clears the convergence bar for a new short (thesis 64, win_prob 72, bear-aligned). Momentum percentile 87 confirms the tape is doing the work, and the AI-layoffs catalyst is directly on-thesis. Risk quality is weak and bounce risk is real after a 65% drop, but the forward setup for further decline is solid. Sizer can handle the late-entry concern.
Better bear convergence than ACN with thesis 64 and win_prob 72, and catalyst confirms AI displacement narrative. But stock has already collapsed 65% and risk_quality is poor — entering a short here carries severe bounce risk. Watchlist skip; not a fresh entry.
Reasonable bear conviction on the structural AI-services disruption narrative, but the stock is already near 52wk lows and PE 13.7 provides some valuation floor. Convergence bar for a new short isn't cleared — thesis and win-prob don't both clear the higher hurdle and most of the easy downside has been captured.
Bear thesis on offshore IT services has real conviction with named AI-displacement catalysts, but stock has already fallen 57% and is near 52wk low — most of the disruption is priced in. We don't short on the watchlist absent a sharper edge, and we're not buying a falling knife with a confirmed bear case. Skip.
Decent bear setup with AI-disruption tailwinds and Eastern European cost-arbitrage erosion, but stock already down 57% and near 52wk low. Bounce risk is real and catalyst flow is only indirectly supportive. Not enough edge from here to short, and no reason to go long. Skip.
Reasonable bear thesis on AI displacement of IT services but stock has already cratered 57% and PE 13.7 provides some floor. Catalyst is thematic rather than name-specific. Doesn't clear convergence bar for new short.
Structural AI disruption thesis on IT services is real, but EPAM is already deeply distressed near 52-week lows. Shorting here offers poor risk/reward — much of the bear case is priced. No specific catalyst to force further breakdown. Skip.
Bear thesis is reasonable but stock has already collapsed 57% and trades at PE 13.7. Catalyst is indirect. Not enough convergence to short here — risk/reward asymmetric with bounce potential.
Bear thesis has some merit but win-probability only 55 and the easy money on the short is gone after a 56% drawdown. No catalyst in the news window. Skip.
Bear case is intact but stock has already collapsed 56% from peak with stabilization signs in recent action. No new catalyst landed in the window. Thesis_pct of 64 is decent but win_prob of 55 says the trade may not work from here. Pass.
Decent bear thesis but win_probability under 60 fails convergence bar. Stock already devastated and PE 14 provides floor. Not actionable for new short.
Bear thesis is plausible given AI displacement of software engineering services, but stock has already collapsed 56% and win_probability is only 55. Risk_quality 20 means dangerous to short here. Not enough convergence for new short.
Bear thesis confirmed by TCS-Anthropic and HCLTech disruption catalysts, but win-probability at 55 fails the convergence test. Stock down 56% from highs at 14x PE — much of the bear case priced in. Risk quality 20 reflects asymmetric short risk. Skip.
Bear thesis is decent but win probability only 55 — much of the damage already done. Doesn't clear the convergence bar for a new short. Skip.
Moderate bear thesis but win_probability only 55 and stock already down 56% from highs limits short upside. Risk_quality of 20 reflects geopolitical and execution risk that cuts both ways. Not enough conviction to short or to flip long. Skip.
Bear thesis fails convergence bar — win_prob below 65 and stock already crushed. Risk/reward asymmetric to the downside on short side. Skip.
Catalyst is the strongest in the batch (Anthropic Claude code disclosure directly threatens dev outsourcing), but win_prob only 58 and stock already deeply discounted. Doesn't meet convergence bar.
EPAM bear case is structurally sound — offshore IT services facing AI coding disruption — but stock already collapsed 54% from highs. Win probability only 58 and risk_quality very low. Skip as new short entry.
Bear thesis valid on IT services AI disruption but stock already crushed. Win-prob below 65 convergence bar for new entry. Cheap valuation and weak momentum suggest limited near-term edge. Skip.
Bear thesis directly contradicted by catalyst — EPAM is building multi-agent AI systems with Claude Code, demonstrating adaptation. Stock already down 54% with cheap valuation. This is exactly the wrong type of bear entry. Skip.
Bear thesis on IT services AI disruption is reasonable but stock has already taken the hit (-54% from highs). Cheap valuation provides downside floor. Win-prob=58 and catalyst=55 don't justify late short entry. Skip.
EPAM bear thesis has confirmation but stock is already deeply distressed (-54% from highs) at 14.7x PE. Win-prob only 58 and risk quality very low at 20. Better bear setups exist (RHI, UPWK, TASK). Skip.
Bear thesis but stock already collapsed 54%. Win-prob 58, catalyst 50, no relevant news. Doesn't clear new-short bar. Skip.
Decent bear convergence on thesis and win probability, but stock already collapsed 54% leaving limited remaining downside. Low risk quality and geopolitical complexity make this a pass for short book.
Reasonable bear setup with AI displacement of IT services as core thesis, but stock already collapsed 54% and risk_quality is poor at 20. Catalyst mildly supportive but not direct. Marginal short, not a high-conviction one.
Bear thesis intact with direct catalyst from Wix AI-driven layoffs, but stock already collapsed 54%. Risk quality very low at 20 (geopolitical). Short setup at lows is asymmetrically bad. Skip.
Direct catalyst (TCS-Mistral AI partnership) supports bear thesis on IT services, but stock has collapsed 54% and risk_quality is poor (20). Late to short here. Skip.
Bear thesis with direct catalyst (AI coding/DevSecOps displacing services) supportive. Win_prob 68 and thesis 67 are close to but below convergence bar (need 65/60). Stock already near 52wk low limits incremental bear edge. Borderline — skip as new short, but track.
Reasonable bear thesis on IT services facing AI disruption, but the stock has already collapsed 54% and trades near 52wk lows. Asymmetry for a fresh short entry is poor. Catalyst confirmatory but no near-term breakdown trigger. Pass.
Bear thesis intact on AI services disruption + geopolitical overhang, but no fresh catalyst and stock already collapsed 54%. Win_probability=68 not enough to override the lack of fresh trigger. Skip new short.
Bear case validated by AI services disruption but stock already down 58%. Risk/reward less compelling than fresh setups. Win-prob 72 and thesis 67 clear bar but risk_quality only 20 suggests squeeze risk on a beaten name.
Bear thesis with reasonable convergence (thesis 67, win_prob 72), but stock already down 58% from highs and near 52wk low. Risk_quality 20 reflects geopolitical and structural concerns. Hard to chase a short into the hole. Skip.
EPAM bear thesis has merit with thesis_pct 67 and win_prob 72, but stock already down 58% from highs near 52wk lows. Risk quality is poor at 20. Bear shorts near multi-year lows have asymmetric squeeze risk. Skip new short entry.
Best bear setup of the IT services cluster — geopolitical overhang plus AI disruption narrative, win_prob 72 and performance_pct 77 confirm breakdown trajectory. However, stock already down 58% from highs raises asymmetry concerns and catalyst only 60. Borderline — not a clear new short entry without stronger catalyst convergence.
Decent bear convergence (thesis 67, win_prob 72) but stock already down 58% from highs sitting near 52wk lows. Risk_quality 20 signals execution danger on entry. Catalyst weak at 40. Not a fresh entry point.
Best bear convergence in the batch with thesis+win_prob+catalyst all aligned. Eastern European software outsourcing directly threatened by Anthropic/MSFT AI coding statements. However stock already crushed -58%, near 52wk low — chasing late shorts is dangerous. Skip new short here but keep on watchlist for bounce-to-short setup.
Bear thesis converging — thesis 67, win-prob 72, stock at $93 near lows with AI services disruption and geopolitical headwinds. However risk_quality 20 signals high volatility and squeeze risk on short side. Doesn't clear new-buy convergence bar (need thesis>=60 and win-prob>=65 — does, but bear shorts get extra scrutiny). Borderline; monitor for cleaner entry.
Strong bear setup with technical breakdown and direct catalyst (IBM AI pods displacing IT services). However, this is a watchlist bear-thesis short, not a long buy candidate. Our mandate favors long convergence trades; shorting at 52wk low after a 55% decline carries asymmetric squeeze risk. Skip as new position.
Strong bear setup: stock at 52wk low, down 55% from high, AI automation directly threatens IT services model. Thesis_pct 67, win_prob 68, catalyst 65 all aligned bearish. However, this is a watchlist bear idea, not a held position, and shorting at 52wk lows carries bounce/squeeze risk. Decent bear conviction within batch but not actionable as new short here.
Strong bear thesis with technical breakdown and AI-displacement catalyst directly hitting EPAM's software services model. High conviction on the bearish view, but as a long-only watchlist name this translates to skip. Ranked highly within batch as a conviction-of-avoidance signal.
Bear thesis on IT services facing AI automation. Stock breaking down to 52wk lows confirms thesis but no direct catalyst and risk_quality is low. Not actionable as a long, and we don't short in this fund. Skip.