Gaurav Kumar.
Building full-stack systems and AI tooling.
Software engineer specialising in backend systems and LLM integration — building enterprise APIs, VS Code extension tooling, and AI-powered features on top of GPT and Claude.
A note on the work.
I build the parts of products that have to be durable — backends that handle money, extensions that handle developer trust, and AI features that have to actually work when the stream stalls or the tool-call returns the wrong shape.
At PureCode AI I lead backend architecture for the enterprise service, engineer an LLM abstraction layer across GPT and Claude variants, and ship agentic tool-calls and a terminal manager into the VS Code extension. I also moved an entire payments stack from Paddle to Stripe — customers, active subscriptions, methods — without a minute of downtime.
Before that, at LearnLater I built LLM streaming, full-text search with Meilisearch, Redis caching, and bidirectional calendar integrations for an enterprise learning platform. Earlier still, I was a backend intern at Massive Mobility, where I built secure APIs and a real-time payments dashboard.
The working stack.
Tools are means, not identities — but a clear inventory is honest. These are the ones I reach for first.
Languages
Backend
Databases
AI / ML
Cloud & DevOps
Where the work happened.
PureCode AI
Backend and AI-tooling work across a VS Code extension, an enterprise service, and a payments stack — the kind of work that has to keep running while it changes underneath.
- Led backend architecture for PureCode Enterprise, defining the Node.js/NestJS/PostgreSQL stack and API design standards adopted across the platform.
- Engineered an LLM abstraction layer integrating GPT-4, GPT-3.5, and Claude variants into the VS Code extension, enabling dynamic model switching and A/B testing.
- Shipped a Terminal Manager and a suite of tool calls (file ops, command execution, output parsing) that expanded the extension's agentic capabilities.
- Built a team-shared rules and prompt library enabling admins to version and distribute reusable LLM instructions, improving response consistency across teams.
- Migrated the payment gateway from Paddle to Stripe (customers, active subscriptions, payment methods) with zero downtime.
- Architected a sales automation pipeline integrating Apollo, HubSpot, and Instantly, automating the full lead lifecycle from extraction to sequenced outreach.
- Containerized services with Docker and built CI/CD pipelines on Azure App Service and AKS, and provisioned staging environments via IaC to eliminate environment drift.
LearnLater
Full-stack engineering on an enterprise learning platform — LLM streaming, search, calendar integrations, and caching that reduced latency at scale.
- Implemented response streaming on LearnUpAI, sharply reducing perceived LLM latency and improving end-user responsiveness.
- Built a tailored learning-plan engine for enterprise clients with automated content recommendations and rules-based plan population.
- Delivered platform-wide full-text search using Meilisearch, unifying discovery across notes, articles, and saved items.
- Built bidirectional Outlook and Google Calendar integrations to create and sync reading events across platforms.
- Developed an email-ingestion webhook that parses messages and auto-associates content with user accounts, replacing manual assignment.
- Implemented Redis and HTTP caching (ETags, Cache-Control) on high-traffic endpoints, improving response times and reducing upstream load.
Massive Mobility
Backend foundations for the Zecat and Bhamo apps, plus a payments dashboard that the team still reads daily.
- Built and deployed secure REST APIs for the Zecat and Bhamo web apps using JWT auth, input validation, and rate limiting.
- Developed a real-time payment analytics dashboard, improving financial visibility for operations and finance teams.
- Strengthened logging and exception handling across services, enabling faster debugging and smoother production issue resolution.
Things I’ve shipped.
Hotinbox Sales Agent
An AI-driven sales workflow that wires Apollo, Clay, HubSpot and Instantly into a single automated lane — list building and enrichment on one end, multi-step outreach and CRM sync on the other. Built for sales teams who want pipeline, not plumbing.
- 01List building + enrichment pipelines across Apollo and Clay.
- 02n8n / Make-style automations syncing leads → CRM → outreach.
- 03Role-based access, campaign analytics, and response-rate dashboards.
PureCode AI · Terminal Manager
A terminal manager inside the PureCode VS Code extension — centralises command execution and terminal state so tool-calls can run, read back output, and stay coherent across sessions.
- 01Single source of truth for command execution across the extension.
- 02Tool-calls for file creation, shell execution, terminal reads.
- 03Reduced flakiness in AI-driven dev workflows.
VoiceType · Whisper-powered Dictation
A macOS menu-bar dictation app — push-to-talk hold-Control, on-device Whisper transcription, then injected straight into the focused field. A local-first clone of WhisperType / Wispr Flow.
- 01100% on-device transcription via pywhispercpp — no cloud round-trip.
- 02Voice-activity detection, hotkey capture, and overlay UI.
- 03Packaged as a signed .app with mic + accessibility entitlements.
Voxa · Multilingual TTS & Voice Cloning
A production-grade text-to-speech web app — 9 languages, voice cloning from a 6–30 s sample, auto-detected CUDA / Apple Silicon MPS / CPU backends. Model loaded once at startup for sub-second generation.
- 01XTTS-v2 voice cloning with custom speaker references.
- 02FastAPI routes for generate / history / download / clone-voice.
- 03Next.js App Router frontend with streaming playback + history.
Paddle → Stripe Migration
A full payment-gateway migration at PureCode — customers, active subscriptions, and payment methods moved with zero downtime for paying users.
- 01Mapped + migrated subscription state without a billing gap.
- 02Reconciliation pipeline to catch drift between providers.
- 03Cutover plan that kept charges flowing during the switch.
Let’s build something.
Open to roles, contracts, and the occasional weekend project — especially anything sitting at the line between product and AI. Reply times are generally within a day.