US 12,443,620 B2
Copilot architecture: network of microservices including specialized machine learning tools
Elaine Kelsey, Corvallis, OR (US); Elliot Nicholas Robson, Seoul (KR); Sazzad Mahmud Nasir, Muncie, IN (US); Jeffrey Thomas Yarbro, Memphis, TN (US); Robert Oscar Robson, Corvallis, OR (US); Lauren Elizabeth Egerton, New York, NY (US); Spencer Thomas Ward, Kent, WA (US); and Brendan Michael Kelly, Somerville, MA (US)
Assigned to THIA ST CO., Corvallis, OR (US)
Filed by THIA ST Co., Corvallis, OR (US)
Filed on Jan. 28, 2025, as Appl. No. 19/039,734.
Application 19/039,734 is a division of application No. 18/898,502, filed on Sep. 26, 2024, granted, now 12,242,503.
Claims priority of provisional application 63/561,654, filed on Mar. 5, 2024.
Claims priority of provisional application 63/620,329, filed on Jan. 12, 2024.
Prior Publication US 2025/0231960 A1, Jul. 17, 2025
Int. Cl. G06F 16/2455 (2019.01); G06F 16/25 (2019.01)
CPC G06F 16/258 (2019.01) [G06F 16/2455 (2019.01)] 24 Claims
OG exemplary drawing
 
1. One or more computer readable media storing instructions executable by one or more hardware processors, the instructions comprising:
a plurality of modules which, when executed by the one or more hardware processors, implement respective microservices, the microservices forming a weakly connected network of microservices configured as a copilot to one or more first client applications;
wherein each of the microservices is configured to:
receive input from (i) a respective first group comprising one or more others of the microservices or (ii) one or more second client applications; and
transmit output to (i) a second group comprising one or more of the microservices or (ii) one or more third client applications;
wherein the network of microservices comprises at least an expansion microservice, a retrieval microservice, one or more core microservices, and one or more evaluation microservices;
wherein, responsive to first input, comprising one or more first tokens, received from the one or more second client applications, the expansion microservice is configured to: determine one or more second tokens associated with but distinct from the first tokens, and transmit the one or more second tokens toward the one or more core microservices;
wherein the retrieval microservice is configured to perform retrieval augmented generation (RAG) by retrieving one or more data objects related to at least some of the first input and forwarding the retrieved one or more data objects toward the one or more core microservices;
wherein a given one of the one or more core microservices is configured to generate, based at least partly on some of the forwarded one or more data objects, a response to the first input, and to transmit the response toward the one or more third client applications;
wherein the one or more evaluation microservices are configured to perform evaluations of outputs from respective source microservices and, based on the evaluations, conditionally forward the outputs toward respective destinations or conditionally invoke respective destination microservices; and
wherein a plurality of the microservices incorporate respective trained machine learning (ML) tools.