Portrait of Ebinesh A

Ebinesh A

Chennai, India

Fullstack developer

I turn problems into products people use, combining frontend, backend, and applied AI.

3+ years of experience · Software Engineer at HCL GUVI
INSIDER INFO

About me

01 / 06
SELECTED WORK AT HCL GUVI

Built at work.

02 / 06

Production systems, measurable improvements, and the engineering behind them. My work spans learner experiences, business operations, and applied AI.

APPLIED AI · DORA

Giving overlooked sales leads a second chance

20,000+calls audited in production

~6,000 leads reassigned after manager review

Full case study
PERFORMANCE · DORA

Making a slow CRM filter feel instant

2,680 → 5.6 msre-enquiry page query time

~480× faster on the measured query

Full case study
ASSESSMENTS · PLACEMENT PREPARATION

From finding a test to understanding the result

8,249active users on mock-test pages in Q3

25,038 views · #3 most-viewed landing page site-wide

Full case study
B2B PRODUCT · PLACEMENT PREPARATION

Helping colleges see who is placement-ready

~3,200students across ~13 college organisations

Cohort dashboards, readiness filters, and assessment reports

AUTOMATION · CONTENT OPERATIONS

Taking company-prep publishing from days to hours

≥ 2 days → ~2 hoursto create and test one company’s data

133 source documents · 170 generated page JSON files

BEHIND THE WORK
APPLIED AI · DORA

Giving overlooked sales leads a second chance

20,000+calls audited in production

~6,000 leads reassigned after manager review

What I built

At HCL GUVI, I was the main author of DORA’s call-quality module. It checks short calls marked “Not Interested” against topic-specific sales guidelines, rejects findings without matching transcript evidence, and sends flagged calls to managers for a final decision. According to the project’s October 2026 production records, it audited more than 20,000 calls; about 6,000 leads were reassigned after manager review. A flag is a review signal, not a confirmed failure. My work included delayed-transcript handling, retryable CRM handoffs and a 210-call replay harness.

My role

Main author of the call-quality module within a shared CRM.

Engineering & impact

Built topic-specific evaluation rules and rejected findings without matching transcript evidence. Handled out-of-order webhooks, delayed transcripts, retryable CRM handoffs, and a 210-call replay harness. More than 9,000 calls were flagged for review; a flag is a review signal, not a confirmed failure.

BEHIND THE WORK
PERFORMANCE · DORA

Making a slow CRM filter feel instant

2,680 → 5.6 msre-enquiry page query time

~480× faster on the measured query

What I built

At HCL GUVI, I optimized DORA’s re-enquiry filter for about 2.69 million leads. Only around 2% had a re-enquiry date, so I added a partial index for that subset. The measured page query fell from 2,680 ms to 5.6 ms—about 480 times faster. Subtracting the indexed count from a precomputed total also reduced empty-date counts from 1,400–1,900 ms to about 79 ms. These October 2026 project figures are database-query timings, not end-to-end page-load measurements.

My role

Owned the re-enquiry filter and query optimization.

Engineering & impact

Only about 2% of 2.69 million leads had a re-enquiry date. Indexing that subset avoided a full scan. For empty-date counts, subtracting the indexed count from a precomputed total cut the measured time from 1,400–1,900 ms to about 79 ms. These are query timings, not whole-page load measurements.

BEHIND THE WORK
ASSESSMENTS · PLACEMENT PREPARATION

From finding a test to understanding the result

8,249active users on mock-test pages in Q3

25,038 views · #3 most-viewed landing page site-wide

What I built

At HCL GUVI, I owned the Placement Preparation mock-test frontend from discovery and reports through coding assessments, and later contributed backend changes for B2C access and multiple organisations. GA4 recorded 25,038 views from 8,249 active users on mock-test pages during July–September 2026. Those figures describe use of the experience, not growth attributed to a single change. HyreNet supplied the assessment engine, and the original assessment backend was primarily a teammate’s work.

My role

Frontend owner; backend contributor for B2C and multi-org support.

Engineering & impact

Integrated GUVI’s HyreNet assessment engine rather than rebuilding it. Added Markdown and MathML explanations, report retry states, guest discovery, and coding-test autosave. The original assessment backend was primarily a teammate’s work; my later backend contributions enabled separate B2C credentials and guest listings.

BEHIND THE WORK
B2B PRODUCT · PLACEMENT PREPARATION

Helping colleges see who is placement-ready

~3,200students across ~13 college organisations

Cohort dashboards, readiness filters, and assessment reports

What I built

I built and extended the college admin experience so placement teams could manage cohorts, compare MCQ and programming results, and filter students by readiness.

My role

Full-stack contributor; integrated team-defined readiness scores.

Engineering & impact

Built org-admin views, cached cohort statistics in Redis, and surfaced readiness bands with filtering and sorting. The team defined the scoring formula. Student and organisation counts describe the platform’s reach; they are not users I personally acquired. The newer CSV-upload and bulk-email interfaces were not live at the time of this write-up.

  • Qwik
  • PHP
  • MongoDB aggregation
  • Redis
Production DB snapshot · October 2026
BEHIND THE WORK
DEVELOPER EXPERIENCE · CODING PRACTICE

Turning exercises into a place to write and run code

+550%increase in average time spent · 2 → 13 minutes

1,345 live problems · 2,000+ production submissions

What I built

I proposed turning static programming exercises into an interactive coding experience so learners could practise instead of only reading. I researched other practice platforms, designed the experience, and built it across the frontend and backend.

My role

Owned the product idea, competitive research, UI design, and full-stack implementation.

Engineering & impact

Built an in-browser editor with server-side judging, hidden test-case feedback, saved drafts, timers, and submission history. The engine supports 15 judged languages plus HTML, CSS, and React previews. Average time spent increased from 2 minutes to 13 minutes after the change—a 550% increase. The live catalogue contains 1,345 problems, with more than 2,000 code submissions in production.

  • Qwik
  • Ace / Sandpack
  • PHP
  • MongoDB
  • Redis
  • Judge0
Live DB counts · Engagement before/after reported by project owner
BEHIND THE WORK
AUTOMATION · CONTENT OPERATIONS

Taking company-prep publishing from days to hours

≥ 2 days → ~2 hoursto create and test one company’s data

133 source documents · 170 generated page JSON files

What I built

At HCL GUVI, I built a Python pipeline that turns source documents into structured company-preparation pages, validates generated output, and checks published content. In my own workflow, creating and testing one company’s data went from at least two days to about two hours. The repository contained 133 source documents and 170 generated page JSON files. Human review remained part of release. These figures are from my December 2025–September 2026 project records: a before-and-after observation of my workflow, not a team-wide productivity experiment.

My role

Sole author of the content-pipeline repo; frontend integration contributor.

Engineering & impact

Added resumable stages, provider switching, rate-limit handling, JSON repair, and live-page validation. The company-prep site had 19 hubs with GA4 traffic; 14 of those had pipeline output in my repo. The time saving is a before-and-after measurement of my own workflow, not a team-wide productivity estimate.

  • Python
  • Gemini
  • Google Drive API
  • Playwright
  • Qwik
Workflow measurement · Repository output · December 2025–September 2026
More work: daily practice, web performance & early-stage products
DAILY PRACTICE · PRODUCTION

A reason for learners to return

Built scheduled daily MCQ and coding tests with queued scoring and reports. Production records show 2,000+ attempts started and about 1,200 submitted; GA4 recorded 13m 58s average engagement per active user on daily-test pages in Q3 2026.

WORDPRESS · PRODUCTION

A faster foundation for the blog

Rebuilt and maintained the custom blog theme. My recorded before-and-after PageSpeed scores improved from 30–40 to 96–100 following the August 2024 performance work. The blog later served 111,676 views in Q3 2026; that traffic is context, not growth attributed to the rebuild.

LIVE STREAMING · INTERNAL PILOT

Owning the product around managed video

Built the Go backend foundation and first Qwik frontend, then extended GUVI sign-in, guest access, permissions, and audit trails with my team. The MVP reached an internal AWS EC2 staging pilot in about three weeks. Fermion handles video; the platform handles identity and access.

AI RESUME BUILDER · CONTRIBUTION

More dependable resume imports

Added schema-constrained AI extraction for a 12-section resume structure and validation after parsing. Moved resume import behind authenticated backend requests, protecting the AI parsing endpoint from unauthenticated use.

Figures are from project records and analytics documented in October 2026. Each story distinguishes my contribution from the wider team’s work.

PROJECTS

Ideas, Shipped.

03 / 06

A selection of things I’ve built end-to-end.
From the first question to the final deploy.

Click any card to discover the story behind it.

View all projects
EXPERIENCE

Experience

04 / 06
HCL GUVI logo

Software Engineer at HCL GUVI

Chennai, Tamil Nadu
  • Software EngineerSeptember 2025–Present
  • Associate Software EngineerSeptember 2024–August 2025
  • Associate Software Engineer InternJanuary 2024–August 2024
  • DORA · Internal CRM

    End-to-end ownership. Own feature delivery for DORA, an internal CRM managing ~3 million lead records, from requirements and full-stack development to AWS deployment.

  • AI-assisted content publishing

    Faster content preparation. Cut preparation from ~4 hours to ~10 minutes (~96% less preparation time) for suitable batches with AI-assisted publishing workflows, keeping human review and approval in place.

  • AI-assisted sales-call review

    Evidence for sales-call review. Built an AI-assisted workflow for calls under 2 minutes marked “Not Interested,” surfacing relevant evaluation guidance and transcript evidence for managers to review and make the final decision.

  • Placement Preparation

    Primary developer for Placement Preparation. Delivered proctored-test integrations, an interactive coding playground, and queue-based assessment reporting, with sole ownership of substantial parts of the platform.

  • Learning-platform prototype

    Rapid product evaluation. Delivered a customized learning-platform prototype in ~3 days, enabling the product team to evaluate an alternative third-party platform.

  • DORA & Resume Builder

    Mentoring through release. Mentored 2 interns and owned code review, integration, testing, and AWS releases across DORA and Resume Builder.

Jan 2024 — Present2 years 10 months
Mallow Technologies logo

Associate Software Engineer Intern at Mallow Technologies

Coimbatore, Tamil Nadu
  • Ruby on Rails · Proof of concept

    Completed a proof of concept using Ruby on Rails, applying MVC patterns and building practical experience with application structure and implementation.

Aug 2023 — Dec 20235 months
Chalkpiece logo

Board Member at Chalkpiece

Chennai, Tamil Nadu · Remote · Part-time

Joined during college as a UI design intern and grew into library coordination, design education, community volunteering, and now a board role.

  • Board MemberJul 2025 — Present · Part-time · Remote
  • Community VolunteerAug 2023 — Jul 2025 · Part-time · Remote
  • Academic College Design Education CoordinatorJun 2023 — Aug 2023 · Part-time · Remote
  • Chalk Library CoordinatorJul 2022 — Jun 2023 · Part-time · Coimbatore
  • UI DesignerApr 2021 — Jun 2022 · Internship · During college
  • Chalk Libraries

    Coordinated 5 libraries across Tamil Nadu, working with teams in different locations.

  • Design education & mentoring

    Mentored a team drafting a design education syllabus for college students as Academic College Design Education Coordinator.

  • ChalkTalk · Podcast host ↗

    Host of Chalkpiece’s podcast, running for more than 3 years, alongside my community and leadership roles.

Apr 2021 — PresentVolunteer experience
NOTES & ARTICLES

Blogs

05 / 06

I write about the problems I run into, the systems I explore, and the ideas I’m still learning. View all blogs ↗

TECH STACK

Tools

06 / 06

Technologies I’ve used across product work, community projects, and engineering experiments.

  • TypeScript
  • React
  • Python
  • Go
  • PostgreSQL
  • AWS
Explore all technologies

Languages

  • JavaScript
  • TypeScript
  • Python
  • PHP
  • Go
  • Ruby
  • Bash

Frontend

  • React
  • Next.js
  • Astro
  • Qwik
  • HTML5
  • CSS3
  • WordPress

Backend & data

  • FastAPI
  • Django
  • Node.js
  • Express
  • Ruby on Rails
  • PostgreSQL
  • MongoDB
  • Redis
  • Firebase
  • Convex
  • Supabase

Cloud & delivery

  • AWS
  • Google Cloud
  • Docker
  • GitHub Actions

AI, real-time & testing

  • Gemini
  • Socket.IO
  • Playwright