Pranav Sharma

Software Engineer

Hyderabad, Telangana, India4 yrs 3 mos experience

Most Likely To Switch

Key Highlights

Designed QEdgeSync, processing 200-300 tickets daily.
Engineered custom Kubernetes Operator for PCIe cards.
Expert in distributed systems and cloud-native services.

Stackforce AI infers this person is a Backend Engineer specializing in distributed systems and cloud-native infrastructure.

Contact

Skills

Core Skills

Distributed SystemsKubernetesEvent-driven ArchitecturePlatform Engineering

Other Skills

C#.NET FrameworkAWS S3AWS SQS FIFOAWS GuardDutyCloudWatchSNSLambdaGolangOpenShiftFluentdLokiGrafanaGo (Programming Language)Observability & Alerting

About

Backend Platform Engineer at Qualcomm with 3 years of experience designing and building production-grade distributed systems, Kubernetes-based infrastructure, and cloud-native backend services. At Qualcomm, I solely designed and built QEdgeSync — a bidirectional cross-platform sync service between Jira and an internal ticketing system, processing 200–300 tickets/day against a system capacity of ~1K tickets/day on AWS (S3, SQS FIFO, Lambda, CloudWatch, SNS, GuardDuty). I also engineered a custom Kubernetes Operator in Golang to manage the full lifecycle of Qualcomm PCIe accelerator cards, with conflict-safe node labeling, zero-downtime rolling upgrades across Amazon EKS and Red Hat OpenShift, and a structured observability pipeline (Fluentd → Loki → Grafana). Features from this work have shipped to client-facing open-source platforms (EKS-A, OCP). My core strengths lie in distributed systems, event-driven architecture, platform engineering, and observability. I work primarily with Golang, C#/.NET, Python, C++, and Bash, and have hands-on production experience with Kubernetes, Docker, AWS, PostgreSQL, DynamoDB, Redis, Prometheus, and Grafana.

Experience

4 yrs 3 mos

Total Experience

2 yrs 1 mo

Average Tenure

3 yrs

Current Experience

Qualcomm

2 roles

Software Engineer

Jun 2023 – Present · 3 yrs

QEdgeSync – Cross-Platform Integration Service
Owned end-to-end design of a bidirectional Jira↔Trinity sync service (AWS S3 + SQS FIFO), processing 200–300 tickets/day at ~1K/day system capacity.
Implemented idempotent comment sync via SQS FIFO ordered delivery and a per-issue Jira comment ledger tracking synced IDs, preventing duplicate syncs across re-runs.
Enforced zero-trust security on S3-staged attachments using AWS GuardDuty Malware Protection tag validation, blocking infected files across ~1K attachments/day.
Built a two-phase SHA-256 deduplication system (filename/size pre-check + hash comparison) eliminating an infinite re-sync loop on every poll cycle.
Designed an alerting pipeline (CloudWatch → SNS → Lambda) with 5xx/4xx log classification, embedding error context in alerts — reducing MTTD to under 1 minute at under $8/month.
PCIe Accelerator Orchestration – Kubernetes Operator
Designed a custom Kubernetes Operator in Golang (CRDs, controllers, reconciliation loops) managing the full lifecycle of Qualcomm X100 PCIe cards, including flashless boot-up via nbdkit + KMM Operator.
Implemented conflict-safe node labeling API with 409 Conflict re-fetch-and-retry (up to 5×), post-write verification, and always-fresh reads — eliminating stale node object errors.
Built a structured Golang logging module with unique reconciliation IDs, integrated with Fluentd → Loki → Grafana for per-reconciliation log filtering.
Engineered zero-downtime rolling upgrades across Amazon EKS and Red Hat OpenShift (99%+ availability), with automated Bash/Python workflows reducing deployment time by 70%.
Collaborated with CE team engineers through code reviews, incorporating improvement PRs and implementing additional performance optimizations on the operator codebase.

C#.NET FrameworkAWS S3AWS SQS FIFOAWS GuardDutyCloudWatch+6

Interim Engineering Intern

Jan 2023 – Jun 2023 · 5 mos

Automated Kubernetes Operator installation without OLM (Operator Lifecycle Manager), reducing manual
steps by 80% for more reliable deployments.
Created a monitoring script to auto-ping the X100 card post-boot, accelerating troubleshooting and improving
operational efficiency.
Improved Operator build tooling by introducing centralized configuration updates, removing the need for repeated manual inputs of repository links, credentials, and parameters.

OpenShiftGo (Programming Language)