Résumé — Phaneendra Bheesetti

Professional Summary

Senior Software Engineer with 6+ years architecting full-stack systems end-to-end, from the data model to the on-call rotation. At Airbus I solo-built ABAcS — a budgeting platform that governs a €500M+ annual budget for 1,500 users across five countries and cut allocation cycles 70% — and ECM, a document-publishing app that eliminated roughly 4,700 hours of specialist manual work. Across six years the constant is order-of-magnitude cycle-time reduction on systems people still depend on daily. IIT Madras alum, now solo-founding KodeRyu, an AI voice interviewer that conducts real spoken technical interviews.

Skills

Languages & Data

TypeScript, JavaScript, Python, Java, SQL, HTML, CSS · PostgreSQL, MongoDB, Redis, Kafka, Google Sheets API

Frameworks & Infra

React, Next.js, Angular, Node.js, Express, FastAPI, Electron, Redux Toolkit, Zustand, Tailwind, Prisma, Django, BullMQ, Docker, nsjail, AWS, Jenkins

AI / Voice / Auth

Claude API, Claude Code, MCP, Deepgram STT, ElevenLabs TTS, Kokoro TTS, Clerk, Razorpay, OAuth / JWT

Testing & Telemetry

Vitest, Playwright, Jest, PostHog, Sentry, GA4, SonarQube, Git, JIRA

Methodology

Agile, Scrum, TDD, CI/CD, Code Review, Incident Response, On-call

Experience

KodeRyu — Solo Founder & Full-Stack Engineer

Jan 2026 – Present

AI voice interviewer · koderyu.com · Feature-complete, heading to launch

Built KodeRyu solo end-to-end — an AI voice interviewer that conducts real spoken technical interviews across voice, DSA and system design — on a React 18 / Express / Postgres monorepo (28 Prisma models, 60+ services, a 502-question bank).
Built a voice engine that holds a natural spoken interview in real time — streaming speech-to-text (Deepgram) into an LLM and back through self-hosted TTS (Kokoro) fast enough to feel like live conversation, recovering when candidates interrupt or go off-script, and giving progressively stronger hints when they're stuck.
Executes untrusted candidate code in 5 languages (Python, JavaScript, Java, C++, Go) inside a Docker + nsjail sandbox on a BullMQ worker pool, with sub-second cold starts.
AI grading that matches human expert graders — κ=0.92 agreement on a 13-criterion rubric — regression-tested by simulating 12 distinct candidate types through an automated evaluation harness.
Shipped the full SaaS solo — Clerk auth, Razorpay tier-gated INR payments, an OWASP Top 10 security pass, DPDP-Act data handling (export + erasure), and PostHog + Sentry + GA4 telemetry.

Tech: React 18, TypeScript, Express, Postgres, Prisma, BullMQ, Kafka, FastAPI, Docker, nsjail, Claude Code

Alten (contracted to Airbus) — Senior Software Engineer

Mar 2021 – Present

Bangalore, India

ABAcS Budgeting Tool

Apr 2023 – Present

Tracks and allocates €500M+ annual budget across 1,500 users (France, Germany, UK, Spain, India)

Solo-architected and delivered the full-stack budgeting platform end-to-end — governing €500M+/yr (€1.5B+ cumulative over 3+ years live) — cutting the allocation cycle 70% and compressing audit prep from days to hours; work conventionally staffed by a 3–5 engineer team.
Designed a multi-tiered role-based access layer adaptable to German, EU and Indian financial regulations, with EU-guideline encryption (CryptoJS) and dashboards tailored to 6 organizational levels.
Integrated Google Sheets as the sanctioned live system of record with a caching layer and global search, keeping a governance-grade platform responsive at scale.
Interviewed users across all 6 organizational levels and shipped a feedback-telemetry loop that accelerated bug resolution — a platform live and audited for 3+ years.

Tech: JavaScript, React, Material UI, Google Apps Script, Google Sheets API, CryptoJS

ECM Tool

Sep 2025 – Mar 2026

Desktop app replacing a legacy Excel-macro workflow for publishing classified documents to the central archive

Architected and shipped the full Electron + React + Node desktop app end-to-end, cutting per-document publishing from 2–3 hours to 10–15 minutes (~90%) — ~4,700 specialist hours eliminated across 2,000+ documents to date (~2.4 person-years).
Implemented OLE container extraction and CFB-library integration to parse binary compound-file structures embedded inside Word documents — deep-systems debugging most workflows never touch.
Built a PDF processing pipeline and a modular plugin (signature removal, metadata extraction, batch processing) — composable workflows that let one publisher sustain a manual team's throughput.

Tech: Electron, React, Node.js, JavaScript, AdmZip, PDF-lib

Quality Tracker Tool (M&P)

Nov 2022 – Feb 2023

Full-stack supply-chain quality tracker with automated alerts and approval workflows

Spearheaded the solo full-stack build in 2 months, delivered to 500+ users across 5 countries, then handed it to a 2–3 engineer team that extended it — solo output that became a product's foundation.
Cut the quality-issue approval cycle from 1–2 days to 2–3 hours (~85–92%) — same-day resolution of issues that used to leave production exposed for days — with zero bug-induced downtime after launch.
Ran user interviews and cross-functional scoping end-to-end before writing code, so the feature set was right the first time.

Tech: JavaScript, HTML, CSS, Google Sheets (DB), Apps Script, AppSheet

MPO (Skywise / Palantir Foundry)

Nov 2021 – Oct 2022

Airbus maintenance-program optimization platform — per-task interval optimization across single-, multi-operator and fleet-wide exercises · 8,000+ operator accounts

Owned the consumption/delivery layer of a platform carrying a multi-million-euro annual cost-avoidance and service-revenue business case — building the pipeline (React-rendered HTML → headless Chromium → paginated PDF) that produces the signed PDF dossiers delivering each optimized maintenance program to operators — the revenue-carrying deliverable they receive.
Built the task-report analysis dashboard front-end — the graphs, tables, filters and range-selection views operators use to explore fleet reliability data (task reports by interval, by aircraft age, operational interruptions) served from Palantir Foundry.
Hardened reliability with Jest unit and integration coverage and served as the team's incident-review focal, resolving 15+ critical production defects on a live revenue product.

Tech: Node.js, React, Redux, Palantir Foundry, AWS, Jenkins, Jest

Composite Compatibility Tool (CCT)

Mar 2021 – Oct 2021

Web app automating compatibility detection for composite materials in aircraft manufacturing

Built the web app from scratch as a 2-engineer team — owning the UI/UX with global search, ag-Grid tables and Firebase auth — cutting the composite-material testing cycle by 80%.
Supported the backend with PostgreSQL schema design and the APIs powering composite-material compatibility and downstream material-validation flows.

Tech: TypeScript, Angular, Jest, PostgreSQL, ASP.NET

Claim Therapist — Software Engineer

May 2020 – Jul 2020

Healthcare Insurance Claim Assist (B2B / B2C)

Built the initial Django product website and partnered with the Founder CTO on ERP design, data models and end-to-end user flows for TPA, hospital and admin dashboards.

Tech: Django, Python

Société Générale — Software Engineer

Jul 2019 – Mar 2020

MyMetrics — Finance Risk Management Division (60K-line PnL & risk-certification platform)

Single-handedly raised a 60K-line PnL & risk-certification platform from a SonarQube D to A+ over 10 months — disciplined reviews, targeted refactoring and UX fixes on a codebase a global bank's risk analysts depended on.
Built 5 new APIs powering the risk and PnL dashboards and complex ag-Grid tables consumed across SG's finance business lines.

Tech: Angular, TypeScript, HTML, CSS, Java, Postgres, SonarQube

Phaneendra Bheesetti