35 agent packages
3 industries publish packages in this role family.
Install-ready machine-learning-focused OpenClaw agents for industry-1-software-it.
3 industries publish packages in this role family.
Builds anti-fraud detection models so invalid traffic, abusive behavior, and monetization risk can be identified before they distort advertiser outcomes.
Builds advertising bid strategies so marketplace monetization systems can improve auction efficiency, return on spend, and commercial performance.
Builds and applies algorithmic solutions with Python, balancing modeling rigor, production practicality, and measurable business value.
Builds recommendation and search-related algorithmic solutions with Python so marketplace ranking quality and user matching keep improving.
Builds Python speech-recognition moderation models so spoken content in videos and live experiences can be transcribed, screened, and escalated with stronger trust-and-safety accuracy.
Builds bidding and budget allocation models so advertiser spend can be optimized against delivery, performance, and pacing objectives.
Builds causal inference and incrementality methods in Python so experimentation, attribution, and strategy decisions can rely on stronger evidence of impact.
Builds computer vision systems with Python, improving image or video understanding, evaluation quality, and production integration readiness.
Builds Python quality-detection models so low-value, spammy, or degraded content can be identified earlier and routed into better ranking, moderation, and ecosystem-quality controls.
Builds Python content-understanding and interest-modeling systems so recommendation, personalization, and creator-user matching improve with richer semantic signals.
Builds content-understanding models and features so marketplace recommendation, search, and governance systems can interpret content more effectively.
Builds conversion attribution models so advertisers and platform teams can interpret performance impact across channels and touchpoints more reliably.
Builds Python estimation models so ad ranking and delivery systems can predict click-through and conversion outcomes with stronger precision.
Operates data labeling workflows that improve annotation quality, throughput, and feedback loops for machine learning model development.
Builds embedding and vector retrieval systems in Python so content understanding, retrieval quality, and semantic matching become stronger across discovery flows.
Builds exploration and exploitation algorithms in Python so distribution systems can learn faster while controlling risk and preserving recommendation quality.
Builds feature platform capabilities with Python and Scala so training and serving workflows use consistent, reusable, and trustworthy feature data.
Leads machine learning strategy, technical direction, and team execution for practical model delivery aligned with product and business outcomes.
Builds Python computer-vision and multimodal understanding models so image and video content can be interpreted more accurately across moderation, risk, and quality workflows.
Builds forecasting models so marketplace inventory and fulfillment teams can plan supply, capacity, and service tradeoffs with better forward visibility.
Builds LLM-powered applications with Python, improving prompt workflows, evaluation quality, and production usefulness for real user tasks.
Builds and productionizes machine learning systems with Python, improving model reliability, deployment quality, and operational usefulness.
Builds model training platform capabilities with Python, improving experiment execution, reproducibility, and scalable model development workflows.
Builds multi-objective optimization methods in Python so ranking and recommendation systems can balance growth, quality, monetization, and risk constraints more coherently.
Builds natural language processing systems with Python, focusing on robust language understanding, evaluation quality, and practical product integration.
Builds Python OCR moderation models so text embedded in images, video frames, and creative assets can be extracted and screened for policy risk more reliably.
Builds marketplace optimization algorithms with Python so subsidy, pricing, and efficiency decisions can improve growth while staying economically disciplined.
Designs prompt systems and evaluation workflows with Python so LLM applications can be improved systematically and measured with discipline.
Builds query-understanding and correction models in Python so search intent parsing, rewrite quality, and user retrieval success improve across content discovery use cases.
Builds Python recommendation and ranking algorithms so personalized distribution quality, user satisfaction, and ecosystem balance improve together.
Builds recommendation and ranking systems so marketplace users receive more relevant products and content with measurable quality gains.
Improves search relevance so marketplace retrieval results better match user intent, catalog structure, and business quality signals.
Builds subsidy optimization strategies so marketplace incentives can balance acquisition, conversion, and unit economics with stronger analytical control.
Builds text relevance and ranking models in Python so search quality, matching precision, and content retrieval relevance improve for diverse query patterns.
Builds Python moderation models for high-risk violative content so policy-breaking political, sexual, violent, and abusive material can be detected with stronger precision and operational coverage.