Pratyush Goel

CEO

Bengaluru, Karnataka, India0 mo experience

AI EnabledAI ML Practitioner

Key Highlights

Led development of AI agent platform enhancing support automation by 60%
Designed and patented a scalable vector database solution
Co-led launch of an advanced question-answering system

Stackforce AI infers this person is a Backend and AI Development specialist in the SaaS industry.

Contact

pratyushgoel99@gmail.com LinkedIn

Skills

Core Skills

Ai DevelopmentBackend DevelopmentSearch TechnologyDatabase ManagementFrontend DevelopmentTeaching And Mentoring

Other Skills

AI AgentsAlgorithmsAmazon DynamodbAmazon S3Amazon SQSAmazon Web Services (AWS)Apache KafkaBixby SDKC++Continuous Integration and Continuous Delivery (CI/CD)Data StructuresDatabase Management System (DBMS)Distributed SystemsDjangoElasticsearch

Experience

Devrev

Member of Technical Staff

Jun 2021 – Present · 4 yrs 9 mos · Bengaluru, Karnataka, India

Led the design and development of an AI agent platform that empowers the creation of custom AI agents, enhancing support automation by 60%.
Tech stack: Python, MongoDB, LangGraph, Langfuse, Temporal, SQS
Designed and developed an intent-aware bot platform to automate customer support, achieving a deflection rate of 20%.
Tech stack: Golang, MongoDB, SQS
Co-led the launch of Turing Answers, an automatic question-answering system leveraging LLM-based Retrieval-Augmented Generation (RAG). The system includes source citations and accuracy analytics, delivering high-quality, reliable answers.
Tech stack: Go, SQS, Semantic Search
Spearheaded the development of a comprehensive semantic search pipeline incorporating indexers, inference systems, rerankers, and feedback loops, leading to a 57% top-3 click-through rate.
Tech stack: Go, Kafka, Vector DB, Embedding Models
Benchmark-tested multiple AI model inference systems in both Go and Python, ultimately selecting TensorFlow Serving to deploy models such as USE and MUSE. This streamlined the deployment process and enhanced model scalability.
Tech stack: TensorFlow Serving
Designed, developed, and patented a highly-available, multi-tenant vector database, capable of scaling to terabytes of embeddings and achieving sub-100ms response times. This innovation resulted in a 50x cost reduction compared to existing solutions like Pinecone.
Tech stack: Golang, Postgres, S3, Redis

PythonMongoDBLangGraphLangfuseTemporalSQS+10

Info edge india ltd

Software Engineer Intern, Platform Team

Jan 2021 – Jun 2021 · 5 mos · Noida, Uttar Pradesh, India

Designed and developed a Java-based framework that automatically generates a GraphQL service over any database by defining the schema in a YAML file.
Built a custom Maven plugin to dynamically generate GraphQL service code, significantly reducing manual development efforts.
Created a reusable library of static code including generic GraphQL resolvers, and established seamless database connectivity for MongoDB, MySQL, and other systems.
Validated the framework by building and deploying multiple internal services for testing and feedback.
Recognized for outstanding performance with a Pre-Placement Offer (PPO) and a Letter of Recommendation.
Tech Stack: Java, GraphQL, Maven, MongoDB, MySQL, YAML

JavaGraphQLMavenMongoDBMySQLYAML+1

Samsung electronics

Software Engineer Intern, Voice Intelligence Team

May 2020 – Jul 2020 · 2 mos · Bengaluru, Karnataka, India

Designed and developed a Bixby voice assistant capsule for Samsung IoT devices, enabling users to control various devices via voice commands.
Implemented backend functionality using Django and MySQL, and built the frontend using the Bixby SDK and JavaScript.
Created two intelligent conversation drivers based on user behavior, enhancing personalization and engagement.
Integrated custom handlers and maintained conversation history for predictive responses.
Secured a Pre-Placement Offer (PPO) for outstanding performance.
Tech Stack: JavaScript, Django, MySQL, Bixby SDK

JavaScriptDjangoMySQLBixby SDKFrontend Development

Coding ninjas india

Intern, Teaching Assistant

Dec 2018 – Apr 2019 · 4 mos · New Delhi Area, India

Mentored over 100 students by delivering optimal and efficient solutions to a wide range of algorithmic problems using C++.
Conducted interactive doubt-clearing sessions for student batches, fostering conceptual understanding and problem-solving skills.
Assisted in debugging and optimizing student code across various data structure and algorithm challenges.
Supported students in completing assignments and projects, ensuring clarity and confidence in implementation.

C++Teaching and Mentoring