Pratyaksh Mundra profile photo 1 on Tal

Pratyaksh Mundra

AI Engineer | GenAI & Agentic Systems | Claude Certified Architect | AWS | ACM MobiCom 2024 | 3x AWS Certified

Bengaluru

5+ years of work experience

HCLTech
Sep 20242 yrs 2 mosPresent

Bengaluru, Karnataka, India

Technology specialist at Commonwealth Bank of Australia ● Led AI and data engineering initiatives for SpendTrack, the bank’s enterprise platform for analyzing personal and business credit card spending across $12B+ in loan exposure, with direct analytics ownership of $2B+ in mapped exposure. ● Developed a GenAI-powered LangGraph agentic workflow that autonomously retrieves and analyzes data from Spark tables, applies regex-driven entity parsing, and generates analytical reports—reducing manual profiling cycles from several days to under 15 minutes and enabling discovery of previously unmapped company exposures. ● Engineered an AWS-based architecture for SpendTrack using EMR, Athena, Step Functions, S3, and IAM, with automated ingestion pipelines (S3 → Spark → Athena) and CloudFormaƟon-based infrastructure provisioning for scalable, production-ready analytics. ● Built PySpark fuzzy matching and suburb classification algorithms to map transactions to precise merchant locations, reducing match time to under 5 minutes per company and boosting accuracy for AI/ML risk models. ● Migrated high-volume SpendTrack workloads to AWS, achieving 50%+ performance improvement in credit risk analytics.
UC San Diego
Jul 20213 yrs 1 moJul 2024

India

Designed and developed real-time location-tracking Android app using Kotlin and Android Studio, seamlessly integrated with AWS for processing, demonstrating proficiency in full-stack app development, cloud computing, and problem-solving. Engineered AWS solution to deploy large-scale IoT networks, focusing on scalability, resilience, and security, demonstrating expertise in AWS services like EC2, S3, Lambda, IAM, CloudTrail, CloudWatch and DynamoDB. Deployed end-to-end pipeline for research and end-user applications involving collecting data to processing and storage in DynamoDB. Achieved latency of around 300 milliseconds. Reduced end-to-end transfer latency by factor of 10 through codebase optimization, improving system efficiency and performance. Developed comprehensive analytics by visualizing of pipeline using Python and Matplotlib, ensuring smooth operations and facilitating decision-making.

Pratyaksh’s AI Stack

claude code
claude code
cursor
cursor
langgraph
langgraph
langsmith
langsmith
mcp
mcp
ollama
ollama

Education

Vellore Institute of Technology

Bachelor of Technology - BTech, Computer Science with sp. In AI and Robotics

Vellore Institute of Technology

20204 yrs2024
Delhi Public School - India

High School Diploma

Delhi Public School - India

20164 yrs2020

Join Pratyaksh and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Pratyaksh Mundra on Tal

Headline

AI Engineer | GenAI & Agentic Systems | Claude Certified Architect | AWS | ACM MobiCom 2024 | 3x AWS Certified

About

AI & Cloud Engineer specializing in scalable ML solutions, GenAI agents, and cloud-based analytics. Experienced in AWS infrastructure, LLM-driven automation, and applied ML for credit risk, IoT, and computer vision. Published researcher (ACM MobiCom ’24) with hands-on expertise bridging AI research and real-world data engineering.

Location

Bengaluru

AI stack

claude code

cursor

langgraph

langsmith

mcp

ollama

Work experience

Senior Consultant at HCLTech, Sep 2024 - Present · 2 yrs 2 mos, Bengaluru, Karnataka, India, Technology specialist at Commonwealth Bank of Australia ● Led AI and data engineering initiatives for SpendTrack, the bank’s enterprise platform for analyzing personal and business credit card spending across $12B+ in loan exposure, with direct analytics ownership of $2B+ in mapped exposure. ● Developed a GenAI-powered LangGraph agentic workflow that autonomously retrieves and analyzes data from Spark tables, applies regex-driven entity parsing, and generates analytical reports—reducing manual profiling cycles from several days to under 15 minutes and enabling discovery of previously unmapped company exposures. ● Engineered an AWS-based architecture for SpendTrack using EMR, Athena, Step Functions, S3, and IAM, with automated ingestion pipelines (S3 → Spark → Athena) and CloudFormaƟon-based infrastructure provisioning for scalable, production-ready analytics. ● Built PySpark fuzzy matching and suburb classification algorithms to map transactions to precise merchant locations, reducing match time to under 5 minutes per company and boosting accuracy for AI/ML risk models. ● Migrated high-volume SpendTrack workloads to AWS, achieving 50%+ performance improvement in credit risk analytics.

Reaserch intern at UC San Diego, Jul 2021 - Jul 2024 · 3 yrs 1 mo, India, Designed and developed real-time location-tracking Android app using Kotlin and Android Studio, seamlessly integrated with AWS for processing, demonstrating proficiency in full-stack app development, cloud computing, and problem-solving. Engineered AWS solution to deploy large-scale IoT networks, focusing on scalability, resilience, and security, demonstrating expertise in AWS services like EC2, S3, Lambda, IAM, CloudTrail, CloudWatch and DynamoDB. Deployed end-to-end pipeline for research and end-user applications involving collecting data to processing and storage in DynamoDB. Achieved latency of around 300 milliseconds. Reduced end-to-end transfer latency by factor of 10 through codebase optimization, improving system efficiency and performance. Developed comprehensive analytics by visualizing of pipeline using Python and Matplotlib, ensuring smooth operations and facilitating decision-making.

Education

Bachelor of Technology - BTech, Computer Science with sp. In AI and Robotics, Vellore Institute of Technology

High School Diploma, Delhi Public School - India