Manulife

Platform Engineer

AedoAI Inc. (Aedo)

DevOps and Automation Engineer

Telfer School of Management at the University of Ottawa

Graduate Teaching Assistant (MIS- Management Information System)

Walmart Canada

Associate

Jio Platforms Limited (JPL)

DevOps Engineer

Ramrao Adik Institute of Technology

Research Internship in Electronics and Telecommunication Department from RAIT.

Bharat Sanchar Nigam Limited

BSNL Internship: Specialization in telecom transmission technologies-satellite communication and OFC

Experienced Engineer with 3+ years of experience in DevOps, automation, and optimizing mission-critical deployments. An analytical and data-driven person who believes in the power of making intuitive decisions by leveraging powerful statistics. Worked with multiple firms to solve critical problems and consistently meet deadlines and targets along with delivering high-quality work. 

Graduated from University of Ottawa with Master’s in Engineering Management degree.

Technical Skills
▪ Cloud Platforms: Microsoft Azure, Amazon Web Services 
▪ Build Tools: Ant, Maven, Gradle 
▪ Containerization: Docker, Kubernetes 
▪ CI Tools: Jenkins, Azure DevOps Pipelines. CD Tools: Jenkins, Ansible 
▪ OS: Linux 
▪ Web Testing Framework: Cypress.io 
▪ Scripting: Python, Shell/Bash Scripting, JavaScript 
▪ Database: MYSQL, MongoDB, PostgreSQL 
▪ Ticketing: Azure, JIRA 
▪ Code Quality: SonarQube 
▪ Data Visualization tools: Power BI, Tableau 
▪ Query Language: GraphQL 
▪ Other Tools: Microsoft Excel

I love networking with new people, so feel free to connect with me!

Imagine you’re sending rent through your banking app.

The screen spins. No confirmation. No error.

…just uncertainty

That tiny moment between what you’re expecting and what’s happening is where reliability comes into the picture.

But how do you define “reliable” in a way that everyone (engineering, product, business) understands?

That’s where three concepts matter.

1/ SLIs: This is what users actually experience
e.g., % successful transfers, latency

2/ SLOs: This is how “good enough” it needs to be
e.g., 99.9% success, 95% < 5s

3/ SLAs: This is what we promise externally
e.g., communicate or compensate when we miss

Once SLOs are defined, something useful comes up.

It’s called an error budget - the amount of failure we’re willing to tolerate over time.

This single number should guide key decisions, such as:
→ Ship new features 
→ Slow down & improve stability 
→ Are we investing in the right places?
→ Is this incident actually impacting users?

With these in place, reliability stops being abstract.

Product, engineering, and operations align around the same outcome, which is “every user can do what they came to do”.

#SRE #SLI #SLO #SLA #Reliability

Roshni Desale

Site Reliability Engineer at Manulife

Your API just got hit with 50,000 requests in one minute. Is that a problem?

Maybe. Maybe not. And that uncertainty is exactly why rate limiting matters. Your API traffic suddenly surges, not because something broke, but because multiple clients hit the same endpoint at once. 

Without guardrails, that surge spreads fast: latency spikes, downstream services overload, real users see failures. This is where rate limiting matters, not to block traffic, but to keep the system from tipping over. 

What it protects against:
1/ Any single client overwhelming shared resources
2/ Overall throughput exceeding safe limits
3/ Random failures (you get smart 429 responses with Retry-After headers instead)
4/ Abnormal spikes that hint at abuse or attacks
5/ Untraceable suspicious behavior (rate limiting makes patterns visible)
6/ Unpredictable load on downstream dependencies

In banking, this might look like:
Normal: ~1,000 transfer requests/min per client
Unusual: 50,000/min → investigate and trace

Rate limiting isn't about blocking users. It's about protecting the platform and the trust built on top of it.

Here's the question to ask your team: When was the last time you stress-tested your rate limits? Or reviewed which endpoints even have them?

Because the best time to find out your limits aren't working isn't during an incident.

#SRE #RateLimiting #Reliability

taplio

Roshni Desale's Linkedin Analytics

Roshni Desale

Performance ScoreRoshni Desale's LinkedIn Performance

Roshni Desale's Best Posts (last 30 days)

Ready to grow your LinkedIn presence?

taplio

.css-6ytdur{background-image:linear-gradient(284deg, rgba(39, 43, 72, 1) 0%, rgba(33, 47, 183, 1) 100%);color:transparent;-webkit-background-clip:text;background-clip:text;}Roshni Desale's Linkedin Analytics

Roshni Desale

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Roshni Desale's LinkedIn Performance

Roshni Desale's Best Posts .css-1emoo84{font-size:var(--chakra-fontSizes-md);font-weight:400;color:var(--chakra-colors-gray-600);}(last 30 days)

Ready to grow your LinkedIn presence?

Roshni Desale's Linkedin Analytics

Performance ScoreRoshni Desale's LinkedIn Performance

Roshni Desale's Best Posts (last 30 days)