Tech Experiments
Small prototypes, tests, and ideas that come from building things hands-on.
Hi, my name is Chakshu.
I build, break, and learn with technology.
Making technology feel fun again
Innovator by mistake, lifelong learner, and tech enthusiast exploring curiosity, systems, and useful digital experiments.
What I do
Projects, experiments, and useful things in progress
Career Experience
Roles, responsibilities, and impact
Salesforce (March 2024 - Present)
- Delivered internal tools and workflow improvements with a focus on reliability, speed, and team usability.
- Collaborated across teams to reduce operational friction and improve delivery quality.
- Improved developer and stakeholder workflows through practical automation and system refinements.
Detailed Explanation
Media.net (March 2020 - Feb 2022)
- Built real-time bidders serving 1M+ QPS in a low-latency (25ms) environment, contributing to over .5M monthly revenue.
- Conceptualized, analyzed, built, and validated machine learning and statistical models to improve prediction quality, accuracy, and system performance, with a primary focus on Decision Trees.
- Partnered with multi-disciplinary product and business teams to identify and prioritize new opportunities for system and revenue growth.
- Designed and implemented multiple proof-of-concepts (POCs) to validate high-potential ideas before full-scale rollout.
Detailed Explanation
Medcords (May 2019 - Feb 2020)
- Built and scaled Python backend services that powered critical day-to-day healthcare workflows.
- Engineered reliable data and API flows across Doctor's app, Patient's app, Pharmacy app, and the Operations Console to keep the ecosystem synchronized.
- Product and backend stack: Kotlin, Jetpack Compose, Angular 6/7, React Native Web, PyPy 3, Python 2, Go.
- Cloud and data stack: Kubernetes (Google Cloud), Docker, AWS (S3, DynamoDB, SES, SNS, Lambda), MongoDB Cloud, Elasticsearch, Redis, Aerospike, RDS, Spark, Hadoop.
Detailed Explanation
Learning Ledger
Failures So Far .....
School
- Built features fast and discovered key assumptions were wrong after release.
- Learned to run smaller user checks before committing to full implementation.
- Now I use tighter feedback loops and staged rollouts.
College
- Spent too much time designing ideal architecture before proving core value.
- Learned to start with the minimum useful version and evolve from real usage.
- Now I prioritize clarity and outcomes over premature abstraction.
Career
- Delayed delivery by not asking for feedback early from teammates.
- Learned that faster collaboration usually beats isolated perfection.
- Now I share drafts sooner and iterate in public with the team.
Say hello