Lakshya Gupta profile photo 1 on Tal

Lakshya Gupta

Platform Engineer @Calance | Go | Kubernetes | AWS | System Design | NodeJs

New Delhi

3+ years of work experience

Calance
Jul 20251 yr 2 mosPresent

Gurugram

- Developed and deployed a decentralized IDP authentication microservice, forming the backbone of the Calance ecosystem. - Played a key role in scaling the client’s platform for 10,000+ active users, proactively mitigating performance bottlenecks. - Automated end-to-end CI/CD workflow creation using Go, reducing setup time per client project from hours to minutes. - Optimized RDS performance through detailed metric analysis, reducing query latency by 20% in a cost-constrained environment. - Set up monitoring stack (Grafana, Alloy, Loki, Mimir) across 32 Kubernetes clusters and 45 VMs, with Alloy on each node and Mimir's ingest endpoint secured via basic-auth, eventually reducing downtime incidents. - Led cloud-to-cloud migration of Atcost food delivery platform (~50K+ downloads), reducing infrastructure cost by 40%. - Refactored backend to offload non-vector data from Milvus to RDS. Improving query latency and lowering operational cost for metadata storage by 95%.
LiveSitter
Nov 20231 yr 9 mosJul 2025

United States

- Architected and deployed infrastructure for real-time AI inference across 1000+ camera streams, leveraging Redis Streams for services workload coordination. - Transitioned to layered microservices architecture ensuring fault tolerance while using AWS Spot Instances, reducing inference and frame decoding costs by 40%. - Strategically migrated from Redis Pub/Sub to RabbitMQ, reducing DB operations from ~1,200 to ~300 per frame, eliminating frame drops and significantly improving reliability of the video processing pipeline. - Configured VPC across inference pipeline, ensuring security of components and significantly reducing data transfer cost. - Automated AI model training leveraging multi-core GPU on EC2 Spot Instances, reducing model training cost by 60%.
DevTown
Jun 20235 mosOct 2023

India · Hybrid

- Designed microservices architecture and schema model for large-scale ERP solution, handling operations across 50 stores. - Provisioned sharded MongoDB cluster on AWS to support high write throughput of ~500 writes/sec while maintaining p99 latency of ~0.2s, for centralized event sourcing improving traceability and auditability.
DevTown
Jan 20235 moJun 2023

India · Hybrid

- Achieved 3×faster inter-service communication in distributed cluster by replacing traditional REST-based communication with gRPC. - Built a service layer over PostgreSQL databases to handle microservice coordination and data consistency. - Securely configured CDN with S3 for static content delivery, with signed URLs to reduce unauthorized access and bandwidth cost.
GitHub
@lakshya404stc
YouTubeyoutube.com/@lakshya404stc

Lakshya’s AI Stack

claude code
claude code
langchain
langchain
langgraph
langgraph
langsmith
langsmith
mcp
mcp
mistral
mistral

Education

Lovely Professional University (LPU)

Bachelor of Technology - BTech, Computer Science

Lovely Professional University (LPU)

20224 yrs2026
dav schools network

intermediate, Non-Medical

dav schools network

20202 yrs2022

Join Lakshya and 10k+ other professionals on

Tal Boss

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

Download Tal Boss

Lakshya Gupta on Tal

Headline

Platform Engineer @Calance | Go | Kubernetes | AWS | System Design | NodeJs

About

TL;DR I build systems that aren’t supposed to be possible at 21. --- I’ve spent the last three years working directly with startups where the learning curve never slows down. Instead of following the typical early-career path, I was thrown into real-world engineering chaos, scaling distributed systems, building production infrastructure, deploying AI workloads at scale, and owning entire pipelines end-to-end. This pushed me far beyond the comfort zone of traditional roles. I’ve engineered microservices powering 10,000+ users, deployed AI agents on decentralized compute, built load balancers, designed distributed filesystems, and optimized cloud infrastructure across AWS, Kubernetes, and gRPC ecosystems. I thrive in environments where speed, reliability, and ownership matter. My focus continues to be: high-performance backend engineering, distributed systems, DevOps automation, cloud infrastructure, and scalable AI systems. I’m not chasing one niche - I operate at the intersection of many. That’s what lets me build faster, adapt quicker, and solve problems that most early-career engineers never get exposed to. If it involves building systems at scale, I’m already interested.

Location

New Delhi

AI stack

claude code

langchain

langgraph

langsmith

mcp

mistral

Work experience

Platform Engineer at Calance, Jul 2025 - Present · 1 yr 2 mos, Gurugram, - Developed and deployed a decentralized IDP authentication microservice, forming the backbone of the Calance ecosystem. - Played a key role in scaling the client’s platform for 10,000+ active users, proactively mitigating performance bottlenecks. - Automated end-to-end CI/CD workflow creation using Go, reducing setup time per client project from hours to minutes. - Optimized RDS performance through detailed metric analysis, reducing query latency by 20% in a cost-constrained environment. - Set up monitoring stack (Grafana, Alloy, Loki, Mimir) across 32 Kubernetes clusters and 45 VMs, with Alloy on each node and Mimir's ingest endpoint secured via basic-auth, eventually reducing downtime incidents. - Led cloud-to-cloud migration of Atcost food delivery platform (~50K+ downloads), reducing infrastructure cost by 40%. - Refactored backend to offload non-vector data from Milvus to RDS. Improving query latency and lowering operational cost for metadata storage by 95%.

MLOps | Backend Engineer at LiveSitter, Nov 2023 - Jul 2025 · 1 yr 9 mos, United States, - Architected and deployed infrastructure for real-time AI inference across 1000+ camera streams, leveraging Redis Streams for services workload coordination. - Transitioned to layered microservices architecture ensuring fault tolerance while using AWS Spot Instances, reducing inference and frame decoding costs by 40%. - Strategically migrated from Redis Pub/Sub to RabbitMQ, reducing DB operations from ~1,200 to ~300 per frame, eliminating frame drops and significantly improving reliability of the video processing pipeline. - Configured VPC across inference pipeline, ensuring security of components and significantly reducing data transfer cost. - Automated AI model training leveraging multi-core GPU on EC2 Spot Instances, reducing model training cost by 60%.

AWS | Backend Engineer at DevTown, Jun 2023 - Oct 2023 · 5 mos, India · Hybrid, - Designed microservices architecture and schema model for large-scale ERP solution, handling operations across 50 stores. - Provisioned sharded MongoDB cluster on AWS to support high write throughput of ~500 writes/sec while maintaining p99 latency of ~0.2s, for centralized event sourcing improving traceability and auditability.

Backend Engineer at DevTown, Jan 2023 - Jun 2023 · 5 mo, India · Hybrid, - Achieved 3×faster inter-service communication in distributed cluster by replacing traditional REST-based communication with gRPC. - Built a service layer over PostgreSQL databases to handle microservice coordination and data consistency. - Securely configured CDN with S3 for static content delivery, with signed URLs to reduce unauthorized access and bandwidth cost.

Education

Bachelor of Technology - BTech, Computer Science, Lovely Professional University (LPU)

intermediate, Non-Medical, dav schools network

Public links

github: https://github.com/lakshya404stc

youtube.com: https://youtube.com/@lakshya404stc