pulpix - video recommendation at scale
TRANSCRIPT
About Pulpix
AI startup based in New-York and Paris
100+ sites using our technology in the world
$850,000 seed round raised in the US
Accelerated in Silicon Valley by
About Pulpix
AI startup based in New-York and Paris
100+ sites using our technology in the world
$850,000 seed round raised in the US
Accelerated in Silicon Valley by
About Pulpix
100+ sites using our technology in the world
Accelerated in Silicon Valley by
$850,000 seed round raised in the US
AI startup based in New-York and Paris
About Pulpix
100+ sites using our technology in the world
$850,000 seed round raised in the US
AI startup based in New-York and Paris
Accelerated in Silicon Valley by
Recommendation at ScaleKey figures
10 million videos
Less than 100 ms response time
10 million events a day
More than one billion training events
Recommendation at ScaleKey figures
10 million videos
Less than 100 ms response time
10 million events a day
More than one billion events
Less than 100 ms response time
More than one billion training events
10 million events a day
10 million videos
Recommendation at ScaleKey figures
Less than 100 ms response time
10 million events a day
10 million videos
More than one billion training events
Recommendation at ScaleKey figures
Content-based EngineFirst approach
Speech-to-text
Content ScoreMetadata
Keywords extractionWeighting
Content-based EngineFirst approach
Speech-to-text
Content ScoreMetadata
Keywords extractionWeighting
Recency boost
Collaborative FilteringHow to put it into practice?
• User-based recommendations- Known users only- Not contextual
Collaborative FilteringHow to put it into practice?
• User-based recommendations- Known users only- Not contextual
• Video-based recommendations- For all users- Fully contextual
Hybrid EngineLinear score combination
ContentEngine
CollaborativeEngine
Context
Recency
Global scoreScores Linear
Model
Hybrid EngineNonlinear embedding combination
ContentEngine
CollaborativeEngine
Context
Recency
Scores Candidate videos
Hybrid EngineNonlinear embedding combination
ContentEngine
CollaborativeEngine
Context
Recency
Global score
Scores Candidate videos
NonlinearModel
Features
Flexibility queuing
Scalability sharding
Fault tolerance replication
High throughput replication
Recommendation at ScaleRequirements
Flexibility queuing
Scalability sharding
Fault tolerance replication
High throughput replication
Recommendation at ScaleRequirements
Flexibility queuing
Scalability sharding
Fault tolerance replication
High throughput replication
Recommendation at ScaleRequirements
Flexibility queuing
Scalability sharding
Fault tolerance replication
High throughput replication
Recommendation at ScaleRequirements
• Reinforcement Learning
• Deep Learning:
○ Recommendation
○ Video recognition
What’s next?Our current R&D
@pulpix
PARIS NEW-YORK
124 rue d’Aboukir75002 Paris, France
584 Broadway New York10012 NY, USA
Pulpix
Pulpix Inc.
+33 (0)6 66 15 02 42 +1 (415) 996 4453
www.pulpix.com
Pulpix is [email protected]