Mohammad Waris profile photo 1 on Tal

Mohammad Waris

SSE @ Oleria | Ex Simpl, Classplus

Bengaluru

7+ years of work experience

Oleria
Dec 20258 moAug 2026

Bengaluru, Karnataka, India

Simpl
Jun 20223 y 4 moOct 2025

Bengaluru, Karnataka, India

• Owned and scaled a real-time antifraud platform processing millions of events and 120K+ requests/min using Go, Python/FastAPI, and Java/Flink for heuristic and ML-based fraud detection. • Optimized Kafka consumer architecture, reducing consumer instances by 50%, CPU utilization from 35% to 12%, and doubling throughput while eliminating consumer lag. • Redesigned Flink pipelines for LSTM-based ML model inference, parallelizing I/O and migrating from DynamoDB to MongoDB, reducing infrastructure costs by 40% while improving reliability. • Increased NeptuneDB graph writer throughput 3× through query merging and buffered writes, enabling a database instance downgrade and saving $2K+ per month. • Migrated a latency-sensitive fraud detection service from Flask to FastAPI, improving concurrency, eliminating 90 ms SLA timeouts, and reducing container requirements by 50%. • Developed fraud-control tooling, including a Go/Redis sliding-window rate limiter for scraping mitigation and a PostgreSQL-backed self-service risk-rule platform, reducing engineering dependency. • Modernized fraud analytics and ETL pipelines using Airflow, Databricks, Dask, and Redshift, reducing processing time by 60% and improving scalability.
Classplus
Jun 20193 yJun 2022

Noida, Uttar Pradesh, India

• Built and scaled real-time chat infrastructure handling millions of messages daily, with 35 ms P95 latency, 1,500 req/s, and 100K+ concurrent WebSocket connections using Node.js, Socket.IO, Redis, MongoDB, Pub/Sub, Nginx, and NLB. • Led EC2-to-EKS migration with event-driven autoscaling and workload separation, improving resilience and reducing infrastructure costs by 40%; later contributed to the organization-wide AWS-to-GCP migration. • Built an online test platform end to end using React, Redux, Node.js, MongoDB, and Redis, scaling as the user base grew from thousands to millions. • Architected a centralized notification microservice supporting multiple product channels, improving delivery reliability and simplifying notification delivery across services. • Improved platform observability and cost efficiency by deploying centralized ELK logging and building infrastructure cost analytics, reducing bug-resolution time by 40% and infrastructure spend by 15%.
GitHub
@warisrao

Mohammad’s AI Stack

claude code
claude code
claude
claude
gemini
gemini
geminicli
geminicli
github copilot
github copilot
claudecode
claudecode

Education

Dr. A.P.J. Abdul Kalam Technical University (AKTU), Lucknow

Bachelor of Technology - BTech, Computer Science

Dr. A.P.J. Abdul Kalam Technical University (AKTU), Lucknow

Join Mohammad and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

Mohammad Waris on Tal

Headline

SSE @ Oleria | Ex Simpl, Classplus

About

Senior Software Engineer with 7+ years architecting scalable backend systems, real-time data pipelines, and full stack products across Fintech, EdTech, and Cybersecurity startups, with proven staff level ownership of critical systems end to end.

Location

Bengaluru

AI stack

claude code

claude

gemini

geminicli

github copilot

claudecode

Work experience

Senior Software Engineer at Oleria, Dec 2025 - Aug 2026 · 8 mo, Bengaluru, Karnataka, India

Staff Software Engineer at Simpl, Jun 2022 - Oct 2025 · 3 y 4 mo, Bengaluru, Karnataka, India, • Owned and scaled a real-time antifraud platform processing millions of events and 120K+ requests/min using Go, Python/FastAPI, and Java/Flink for heuristic and ML-based fraud detection. • Optimized Kafka consumer architecture, reducing consumer instances by 50%, CPU utilization from 35% to 12%, and doubling throughput while eliminating consumer lag. • Redesigned Flink pipelines for LSTM-based ML model inference, parallelizing I/O and migrating from DynamoDB to MongoDB, reducing infrastructure costs by 40% while improving reliability. • Increased NeptuneDB graph writer throughput 3× through query merging and buffered writes, enabling a database instance downgrade and saving $2K+ per month. • Migrated a latency-sensitive fraud detection service from Flask to FastAPI, improving concurrency, eliminating 90 ms SLA timeouts, and reducing container requirements by 50%. • Developed fraud-control tooling, including a Go/Redis sliding-window rate limiter for scraping mitigation and a PostgreSQL-backed self-service risk-rule platform, reducing engineering dependency. • Modernized fraud analytics and ETL pipelines using Airflow, Databricks, Dask, and Redshift, reducing processing time by 60% and improving scalability.

Software Engineer at Classplus, Jun 2019 - Jun 2022 · 3 y, Noida, Uttar Pradesh, India, • Built and scaled real-time chat infrastructure handling millions of messages daily, with 35 ms P95 latency, 1,500 req/s, and 100K+ concurrent WebSocket connections using Node.js, Socket.IO, Redis, MongoDB, Pub/Sub, Nginx, and NLB. • Led EC2-to-EKS migration with event-driven autoscaling and workload separation, improving resilience and reducing infrastructure costs by 40%; later contributed to the organization-wide AWS-to-GCP migration. • Built an online test platform end to end using React, Redux, Node.js, MongoDB, and Redis, scaling as the user base grew from thousands to millions. • Architected a centralized notification microservice supporting multiple product channels, improving delivery reliability and simplifying notification delivery across services. • Improved platform observability and cost efficiency by deploying centralized ELK logging and building infrastructure cost analytics, reducing bug-resolution time by 40% and infrastructure spend by 15%.

Education

Bachelor of Technology - BTech, Computer Science, Dr. A.P.J. Abdul Kalam Technical University (AKTU), Lucknow

Public links

github: https://github.com/warisrao