Community Outreach Targets
August 5, 2026 ยท View on GitHub
Use this page to share the project with relevant communities without spam.
Primary Audience
- Quant researchers and students.
- ML engineers interested in tensorized financial panels.
- Research software users who care about reproducibility.
- Portfolio and backtesting tool builders.
Recommended Communities
Post only where you already have an account and can reply to comments. Prefer
the v0.2.1 validation entrypoint release when the audience cares about reproducibility
more than launch novelty.
| Channel | Angle | Suggested ask |
|---|---|---|
| GitHub Issues / Discussions | Project launch and benchmark call | Submit CPU/GPU benchmark results |
| Hugging Face Papers | Paper and reproducibility artifacts | Try the public-data and synthetic reproduction paths |
| Awesome Quant | Curated open-source discovery | Evaluate the documented research stack |
| Hacker News Show HN | Runnable open-source research stack | Run the quick start and critique the design |
| QuantConnect forum | Reproducible factor and backtest research | Compare assumptions and public-data results |
| Research software announcement | Feedback from quant/ML engineers | |
| X / Twitter | Short technical launch | Star, benchmark, or try notebook |
| Quant StackExchange chat / communities | Factor-engine discussion | Edge cases and assumptions |
| University quant clubs | Student-friendly baseline | Run Start Here and report issues |
Reddit communities often remove direct repository promotion. Share there only when a substantive, self-contained technical write-up is allowed by the current community rules; the repository link should support the analysis rather than be the purpose of the post.
Current Live Leads
Use these as warm, high-relevance entry points. Do not repost the same pitch; reply only when the thread context makes the project genuinely useful.
| Lead | Status | Good angle |
|---|---|---|
| JustinSongXh/quant-trading issue #35 | Technical reply posted | Eastmoney sector-constituent provider state, stale snapshot semantics, and deterministic A-share ranking inputs |
| Kaoruha/GinkGO issue #6853 | Technical reply posted | Narrow factor-subsystem wiring slice, Alpha158 registry repair, idempotent storage, and stub visibility |
| minihellboy/factorminer issue #36 | Technical reply posted | Independent public-evidence reproduction report template and negative-reproduction value |
| liuh886/alpha_engine issue #531 | Technical reply posted | Confidence-gate confusion table, regime-conditioned factor negative control, and cost/turnover attribution |
| Berryoatzii/Innovestx-bot issue #3 | Technical reply posted | NO-GO to SHADOW replay artifact, deterministic strategy spec, and live-execution evidence gate |
| Hamrez95/Quantara issue #110 | Technical reply posted | Per-playbook promotion packet across replay, shadow, canary, calibration, latency, and signals/week |
| Jettlin927/Quantitative_trading issue #57 | Technical reply posted | A-share value-quality low-turnover review seams, PIT financial visibility tests, and capacity fail-closed priority |
| tonywxx/adaq issue #29 | Technical reply posted | Research workspace evidence-completeness gate and identity-matching acceptance artifacts |
| magilliam27/MCI-GRU issue #123 | Technical reply posted | PIT-aware graph selection fixture and separate consumed-edge vs selection-budget invariants |
| ganluCoding/etf_t-0_hk_stock issue #1 | Technical reply posted | T+0 ETF grid break-even ledger, conservative OHLC execution tiers, and paper-execution validation |
| minihellboy/factorminer issue #31 | Technical reply posted | Proof-carrying alpha claim lifecycle, evidence-budget feedback resolution, and kernel-owned promotion states |
| Hamrez95/Quantara issue #5 | Technical reply posted | Promotion-basis object, rejected-strategy memory, and parameter-selection stability checks |
| Hamrez95/Quantara issue #52 | Technical reply posted | Strategy-specific promotion gates, regime mismatch diagnostics, playbook coverage, and cost-stress validation |
| Hamrez95/Quantara issue #113 | Follow-up technical reply posted | Per-playbook promotion records for v1.2 champion/challenger evidence across replay, shadow, and canary gates |
| HKUDS/Vibe-Trading issue #969 | Technical reply posted | Evidence-stage strategy discovery registry across Alpha Zoo and SDM with reproducible metrics and stale-evidence refusal |
| SiinXu/stock-pulse-ai issue #210 | Technical reply posted | Portfolio stress scenario provenance, deterministic impact modeling, rebalance audit, and T+1/liquidity edge cases |
| Undermybelt/ict-engine-release issue #272 | Technical reply posted | PIT observation-outcome evidence bundle with ledger, split manifest, metrics, fill policy, and replay command |
| RussellFeinstein/WoW-Economy-Forecaster issue #100 | Technical reply posted | Purge/embargo target-availability invariant, split manifest metadata, and horizon-specific leakage tests |
| vinwiegman/quant issue #1 | Technical reply posted | Walk-forward prediction artifact schema, fold invariants, and cost-aware benchmark reporting |
| Stellar-devs-dashboard issue #33 | Technical reply posted | Auditable first-slice portfolio optimizer, post-solve diagnostics, and safer baseline acceptance criteria |
| alex-place/lantern-os issue #3001 | Technical reply posted | Thin overnight alpha cost ledger, auction slippage breakdown, and net-of-cost promotion gate |
| RussellFeinstein/WoW-Economy-Forecaster issue #16 | Technical reply posted | Fold-quality table, usable-fold gate, skipped-fold reason codes, and target-horizon metadata |
| jjakimoto/research-issues issue #319 | Technical reply posted | A-share model-swap reproduction checklist and validation-first framing for vanishing model advantage |
| JohnGavin/historical issue #624 | Technical reply posted | Gross/net exposure, cost-unit reporting, realized turnover, and borrow-cost visibility for strategy leaderboards |
| sunyxi/SeekAlpha issue #52 | Technical reply posted | Strict StrategyPackage handoff contract for PIT universe, split manifests, cost assumptions, gates, and replay artifacts |
| Luke-Bradford/eBull issue #2282 | Technical reply posted | Research-corpus versus execution-venue boundary, delisting register fields, and symbol/date tradability provenance |
| liuh886/alpha_engine issue #342 | Technical reply posted | CN130 ranking ledger, neutralized Rank IC, leave-one-out stability, and portfolio-return-as-secondary-evidence framing |
| HKUDS/Vibe-Trading issue #887 | External fix acknowledged | Identity gate and tool-batch barrier landed on main with regression coverage after earlier design discussion |
| HKUDS/Vibe-Trading issue #886 | External fix acknowledged | Structured evidence ledger and final-answer market-claim consistency landed on main with regression fixture |
| ml-quant-trading issue #7 | Maintainer update posted | August reproduction challenge reminder and structured reproduction-report template link |
| ml-quant-trading discussion #13 | Maintainer update posted for v0.2.3 | Validation dashboard, AkShare coverage gate, and independent equity_curves.csv reproduction ask |
| FinMamba issue #5 | Technical reply posted | Reproduction bundle contract for financial ML paper results: universe, calendar, feature/label hashes, predictions, seeds, and backtest assumptions |
| mechanic-Q/electric issue #61 | Technical reply posted | Explainable replay evidence snapshot, same-window strategy metrics, and assistant context separation |
| Qlib discussion #2284 | Open discussion about A-share factor labs and AI research reports | Share validation-first design notes and mention the repo as a narrower PyTorch/A-share reference, not as a Qlib replacement |
| awesome-quant-ai PR #43 | Merged directly by maintainer | Keep relationship warm; future follow-up should be about concrete benchmarks or an arena-style demo, not another listing request |
| awesome-math-and-trading issue #3 | Open resource suggestion | Wait for maintainer activity before following up; keep the ask about fit and category placement |
| awesome-fintech PR #50 | Open listing PR | No pressure follow-up only if the maintainer asks for changes or the PR becomes stale |
| awesome-ai-trading PR #4 | Open listing PR | Wait for maintainer review; no follow-up unless requested or stale for several weeks |
| coinbase_futures_bot issue #405 | Technical reply posted | Good validation/replay-manifest discussion; no follow-up unless the maintainer asks |
| ai-trade-flow-platform issue #63 | Technical reply posted | Taiwan-market live-trading boundary discussion; no follow-up unless the maintainer asks |
| MCI-GRU issue #116 | Technical reply posted | PIT universe / benchmark-basis correctness discussion; no follow-up unless the maintainer asks |
| quant-trading-system issue #1 | Technical reply posted | Research-only validation workflow discussion; no follow-up unless the maintainer asks |
| Velocity issue #48 | Technical reply posted | Soft-market archive / CLV validation discussion; no follow-up unless the maintainer asks |
| Sport-Betting-Engine issue #46 | Technical reply posted | Model-change validation tier policy discussion; no follow-up unless the maintainer asks |
| ml-quant-trading discussion #13 | Maintainer updates posted | Public discovery surface, Colab demo fix, and benchmark ask consolidated for visitors |
| tradingV3 issue #323 | Technical reply posted | French backtest UI / proof-contract discussion; no follow-up unless the maintainer asks |
| claude-ai-trading-automation issue #38 | Technical reply posted | Earnings-event backtest boundary and point-in-time event-table design |
| quant-scratch issue #6 | Technical reply posted | Source-agnostic local data cache, replay manifests, and yfinance rate-limit visibility |
| yfinance discussion #2503 | Technical reply posted | Single-ticker rate-limit diagnostics and explicit public-data blocker reporting |
| NASDX issue #45 | Technical reply posted | A-share backtest weight validation, fail-fast contracts, and executed-weight diagnostics |
| PortfolioOptimisers.jl issue #174 | Technical reply posted | Portfolio terminology boundary for assets, factors, features, and similarity matrices |
| lookaheadbench issue #1 | Technical reply posted | Structured finance-context leakage, bitemporal artifacts, and replay-safe factor validation |
| AI-Quant-Research-System issue #1 | Technical reply posted | PIT vendor ingestion contract, synthetic fixtures, and no-redistribution data boundary |
| Algo-Quant-Backtester issue #5 | Technical reply posted | Evidence ladder for quickstart, baselines, costs, bias checks, and README claims |
| ernie55ernie/awesome-quant PR #1 | Merged by maintainer | Added ml-quant-trading to Research Frameworks source data for an auto-refreshed quant list |
| NASDX issue #44 | Technical reply posted | Missing/suspended-price handling, stale valuation prices, and execution-price separation |
| NASDX issue #43 | Technical reply posted | Explicit weekly/monthly rebalance calendar semantics and no-lookahead schedule tests |
| yfinance discussion #2912 | Technical reply posted | EDGAR filing backend provenance, accession IDs, same-day filing collisions, and public-data validation |
| yfinance discussion #2783 | Technical reply posted | Polars migration benchmark contract and golden-output compatibility checks |
| yfinance discussion #2854 | Technical reply posted | Daily-bar freshness diagnostics, exchange timezone checks, and data-vintage discipline |
| Alembic issue #57 | Technical reply posted | SPX VRP option-strip data contract, quote provenance, and evidence-first NO-GO criteria |
| beacon-trader issue #61 | Technical reply posted | Replay-safe market-structure versions and shadow attribution before gating |
| AI-Quant-Portfolio-Manager issue #3 | Technical reply posted | Multi-provider market-data contract for source, timezone, vintage, missingness, and license boundaries |
| AI-Quant-Portfolio-Manager issue #1 | Technical reply posted | Research, paper-trading, and live-trading boundary with independent risk gates |
| ml-quant-trading discussion #13 | Maintainer update posted | Landing update for the validation-focused outreach campaign and reproduction-report ask |
| AlphaGPT issue #1 | Technical reply posted | A-share factor-mining output contract, RankIC diagnostics, and SIGNAL_DIR manifest design |
| historical issue #120 | Technical reply posted | StockGPT replication boundary, PIT universe checks, baselines, and factor attribution |
| digithings issue #396 | Technical reply posted | Quant sandbox deterministic smoke tests and public-data-vs-synthetic validation split |
| AlphaForge issue #8 | Technical reply posted | Constrained Markowitz post-solve audit object and baseline comparison evidence |
| finance-query issue #274 | Technical reply posted | Margin/leverage accounting, sizing schemes, and multi-objective optimization separation |
| algua issue #479 | Technical reply posted | PIT-safe deterministic portfolio construction, correlation pruning, and config hashing |
| Quant-Lab issue #2 | Technical reply posted | A-share roadmap ordering around data, labels, factor validation, walk-forward, and models |
| Trading-engine issue #1243 | Technical reply posted | Predeclared calibration report template and test-set discipline for RSI2 validation |
| librae issue #18 | Technical reply posted | Backtest, simulation, and live-trading mode contracts with broker-fill provenance |
| truealpha issue #429 | Technical reply posted | Fixture-vs-real-data provenance and deployed-output drift detection |
| etoro_nautilus issue #736 | Technical reply posted | Process handoff provenance for orchestrator, live node, and optimization artifacts |
| stellarlend issue #682 | Technical reply posted | Risk-aware lending allocation optimizer with gas, liquidity, and attribution reporting |
| stellarlend issue #666 | Technical reply posted | Auto-compounding break-even rules and gas-vs-yield sensitivity |
| ml-quant-trading discussion #13 | Maintainer update posted | Landing update for the new-project outreach pass and reproduction-report ask |
| External listing PRs | Status checked, no nudges posted | Reviewed open awesome-list PRs and avoided premature maintainer pings |
| twopercent issue #60 | Technical reply posted | Multiple-comparison accounting, promotion-band provenance, and shuffled-label baseline discipline |
| quant_intel_system issue #1 | Technical reply posted | Qlib parallel backtest reconciliation boundary for A-share signals and custom engine comparison |
| skrymer/trading issue #182 | Technical reply posted | Regime recall validation with CRISIS protection scorecard and pre-registered anchor-span reporting |
| ml-quant-trading discussion #13 | Maintainer update posted | Landing update for the fresh-project outreach pass and reproduction-report ask |
| v0.2.1 release | Published | Validation entrypoints, Colab bootstrap fix, and outreach follow-through release |
| awesome-systematic-trading PR #72 | Low-pressure follow-up posted | Mentioned v0.2.1 validation entrypoint release and offered wording/category changes |
| Awesome-pytorch-list PR #177 | Low-pressure follow-up posted | Mentioned v0.2.1 validation entrypoint release and offered wording/category changes |
| awesome-ai-in-finance PR #193 | Low-pressure follow-up posted | Mentioned v0.2.1 validation entrypoint release and offered wording/category changes |
| finverify-llm issue #53 | Technical reply posted | Async yfinance offload with bounded concurrency, timeout, stale-cache, and regression test guidance |
| mars-research-assistant issue #103 | Technical reply posted | Explicit data-source selection, OHLCV quality gates, partial-bar rejection, and chart provenance |
| noctis issue #142 | Technical reply posted | Reporting-only metrics boundary, PSR/DSR inputs, benchmark roster freeze, and null metric handling |
| QuantConnect Lean issue #9638 | Technical reply posted | Option-spread grouping invariants, order-route equivalence, and cash-insolvency assertion |
| tradingTplus issue #99 | Technical reply posted | Daily/intraday feature-flow separation, target-timeframe warm-up, NULL semantics, and legacy-column consumer mapping |
| coinbase_futures_bot issue #569 | Technical reply posted | Funding proxy provenance, sign/rank validation, walk-forward calibration, and timestamp leak controls |
| adaq issue #12 | Technical reply posted | Indicator Plan hashing, slot availability masks, gap segment ids, non-finite fail-closed checks, and external-factor provenance |
Short Post
I released ml-quant-trading v0.2.1, an end-to-end PyTorch stack for ML multi-factor research.
It includes 213 factors, masked tensor ops, bias correction, MLP/Transformer baselines, Markowitz optimization, vectorized backtesting, a public-data notebook, CI, and benchmark tooling.
Repo: https://github.com/initial-d/ml-quant-trading Release: https://github.com/initial-d/ml-quant-trading/releases/tag/v0.2.1 Paper: https://arxiv.org/abs/2507.07107
I am looking for CPU/GPU benchmark results and public-data reproduction feedback.
Longer Community Post
I released ml-quant-trading v0.2.1, a research-oriented implementation of
ML-enhanced multi-factor quantitative trading.
The goal is to provide a reproducible baseline rather than a trading signal claim. The repo contains:
- 213 factor dimensions
- mask-aware PyTorch tensor primitives
- limit-up / limit-down / halt bias correction
- MLP and Transformer baselines
- Markowitz portfolio construction
- vectorized backtesting and metrics
- synthetic and public-data demos
- public validation digest and Baostock A-share validation report
- CI, tests, benchmark scripts, and contribution templates
The paper used proprietary data, so the public repo focuses on synthetic and public-data reproduction paths. I would especially appreciate:
- CPU/GPU benchmark reports
- public-data case studies
- factor-engine edge cases
- documentation and setup feedback
Repo: https://github.com/initial-d/ml-quant-trading Release: https://github.com/initial-d/ml-quant-trading/releases/tag/v0.2.1 Paper: https://arxiv.org/abs/2507.07107
Posting Rules
- Customize the first paragraph for each community.
- Disclose limitations clearly.
- Do not imply live trading profitability.
- Do not ask directly for stars before asking for useful feedback.
- Verify the current community rules before posting, especially for self-promotion.
- Reply to comments and convert repeated questions into docs.