Skip to content
Akansha Tiwari

Hi, I'm Akansha.

I make complex systems legible to the people who run them.

I'm a full-stack software engineer going deeper into backend and distributed systems. I added a distributed cache to Thanos, a CNCF project, in Go, and spent four years at Phaidra building tools over thousands of plant sensors.

akanshat1999@gmail.comCVLinkedInGitHub
Akansha smiling in front of tropical plants
Open to workFull-stack and backend roles
Remote or hybrid
IST in India

  1. Before

    Four years of frontend at Phaidra

    First frontend hire at an NVIDIA-backed startup. I built the tools engineers used to map thousands of plant sensors and read charts at a million points per series.

  2. Nowupdated Sep 2026

    Going deeper on distributed systems

    Reading Designing Data-Intensive Applications and working through Fly.io’s Gossip Glomers challenges in Go.

    akanshat/maelstrom-engine ↗

Case studies

Three problems I worked on, each written up in full.

LFX Mentorship with Thanos (CNCF) · Go

Groupcache for Thanos Store

The problem

When many queries asked Thanos Store for the same object at once, each one went to object storage.

What I did

Added groupcache, later moved to galaxycache, as a cache backend in Go, with Prometheus metrics and end-to-end tests.

What changed

Thanos users can cache across Store peers in-process, with no separate cache service to run.

queryqueryqueryquerygroupcache1 tripobject storage4 queries
Read the case study

Phaidra · React · TypeScript

Mapping plant data at scale

The problem

Engineers had to map a plant’s data with no spec, and plants grew to thousands of components.

What I did

Owned the product and UI. Virtualized the heaviest views, paginated the tables, and moved a 10,000+ item picker onto a paginated search API.

What changed

A tool that stayed fast at plant scale, built so people who don’t write code could use it.

⌕ search 10,000+ componentsPlantChiller plantChiller 1Pump 1APump 1BChiller 2Cooling towerFan 3Sensor T-104only visiblerows are drawnsearch APIpaginated1,000sof componentsper plant
Read the case study

Phaidra · React · AG Charts

Charts at a million points

The problem

An AI feature needed to answer questions with charts of six or more series, around a million points each.

What I did

Compared charting libraries, chose AG Charts over Highcharts, and built the charts.

What changed

Charts that hold about a million points per series inside the product.

6+ series× ~1,000,000 points eachpicked AG Charts over Highcharts for this
Read the case study

Where I’ve worked

Four years of frontend on a product team, and backend work in Go on a CNCF project.

Download CV

  1. Phaidra

    Remote

    NVIDIA-backed startup building AI agents for data-center operations. I worked on the frontend.

    Software Engineer IIpromoted

    Jul 2023 – Jan 2026

    • Chose AG Charts over Highcharts for an AI feature that answers questions with charts of six or more series at around a million points each, and built its charts.
    • Built versioned editing for part of the product’s control side, so users could change, review and roll back their settings safely.
    • Wrote PRDs, which were rare at Phaidra, and test plans for my features; reviewed backend API specs and worked with design on flows.
    • Reviewed code, fixed broken pipelines and tests, took on-call shifts, and helped QA engineers get their Cypress tests running.

    Software Engineer I

    Dec 2021 – Jul 2023

    • Joined as the first frontend hire. Reviewed and refactored contractor code, then rewrote the main chart in the customer portal.
    • Owned an internal data-mapping tool for solution engineers, from an open brief to production, designed so people who don’t write code could use it.
    • Kept it fast as plants grew to thousands of components: virtualized the heaviest views, paginated the tables, and moved a 10,000+ item picker onto a paginated search API, with pagination and filters kept in the URL.

  2. LFX Mentorship with Thanos (CNCF)

    Remote · open source

    Open-source, highly available Prometheus with long-term storage. A Cloud Native Computing Foundation project.

    Mentee

    Sep 2021 – Nov 2021

    • Added groupcache, later migrated to galaxycache, as an embedded distributed cache backend for Thanos Store's caching bucket in Go (PR #4818, 26 commits).
    • Worked through review: moved to galaxycache, added Prometheus metrics and end-to-end tests. Released in Thanos v0.25.

  3. Reliance Jio

    Mumbai, India

    India’s largest telecom operator. I worked on JioMart’s B2B partner platform.

    Software Engineer

    Jul 2021 – Nov 2021

    • Led the migration of the JioMart B2B partner app from native Android to a responsive web app, keeping full functional parity across platforms.
    • Shipped production features for a large-scale B2B partner onboarding platform, set up the product's design system, and wrote its tests.

  4. Rakuten

    Bangalore, India

    Global e-commerce and internet services company.

    Software Engineer Intern

    Jan 2021 – Jun 2021

    • Built an internal testing tool that let QA engineers define automated test plans and test steps.
    • Took part in product requirement planning and design reviews alongside the full-time team.

What I work with

Backend & data

  • Node.js
  • Go
  • REST APIs
  • PostgreSQL
  • MongoDB
  • Prometheus

Frontend

  • TypeScript
  • React
  • Next.js
  • AG Charts
  • Redux
  • Tailwind CSS
  • Material UI
  • Sass
  • Vite
  • Webpack

Quality & delivery

  • Playwright
  • Unit testing
  • CI/CD
  • Docker

AI in the workflow

  • Cursor
  • Claude
  • LangChain
  • RAG

Education

B.Tech, Computer Science

IIIT Gwalior · 2017 – 2021 · CGPA 8.08 / 10

Data structures & algorithms · Operating systems · Computer networks · Systems programming

Say hello

If you're hiring for a full-stack, backend or distributed systems role, I'd like to hear from you. Email is the quickest way to reach me.

akanshat1999@gmail.comCVLinkedInGitHub
esc
↑↓ to move↵ to select