LK Lakshya Khanna
open to work --:--

Lakshya Khanna · Full-Stack Engineer · available for 2026

Software that watches, listens & holds the line.

Currently

Contracting with CS Technology on Ferroglobe (US) — monitoring & AI tooling for industrial plants that can't afford downtime.

Based in

Dehradun, India
30°N · 78°E · IST

01

About

profile
LK N 30°19′ E 78°02′ EST. 2016 FULL-STACK

I'm Lakshya — a full-stack engineer who gravitates toward the unglamorous seams of systems: heartbeat agents pinging plant infrastructure, Whisper pipelines chewing through audio, log analyzers that read their own stack traces.

Seven years moving between front-end and back-end work, chasing one kind of problem: systems that need a careful pair of hands to stay upright. I'm happiest when the work is quiet, durable, and slightly boring — when a tool I wrote three months ago is still running and nobody had to think about it.

0+ Years writing software
0h Whisper fine-tune audio
0 Open-source tools shipped
02

Selected Work

one at work · two open-source
MES Assist
furnace temps last 24h?
QUERYING SQL…
FURNACE 01 · LAST 24H
Ask anything…
<script src="/chatbot-widget.js" />
Work / 01

A chatbot that reads docs, runs SQL, and answers in tables and charts.

MES Assist — configurable AI assistant + embeddable widget.

Built end-to-end for a Manufacturing Execution System — a no-code admin panel to configure chatbots (prompts, model, RAG knobs) and a self-contained widget that drops into any page with one <script> tag, isolated by Shadow DOM. LlamaIndex + Qdrant RAG with a data-agent running read-only SQL against live plant data — answers come back as tables and 11 chart types. LLM provider switchable per chatbot.

DjangoDRFReactTypeScriptLlamaIndexQdrantDockernginxAzure OpenAI
In production at CS Technology
log-analyzer.py

$ tail -f /var/log/api.log

⚠ TypeError at line 142 → req.user.id (undefined)

✓ trace captured · context loaded (3 files)

→ proposing fix…

+ if (req.user?.id) return user.id;

- return req.user.id;

✓ slack.notify() → #alerts

$ ▌

OSS / 01

An agent that reads its own stack traces.

Open-source AI log analyzer & auto-debugger.

A Python agent that watches multi-service apps in real time and catches crashes the moment they happen. When something breaks, an AI pipeline reads the stack, locates the seam, and proposes an immediate code fix — front-end or back-end. Slack and email pings carry the diff, not just the error.

PythonOpenAIFastAPISlack APISMTP
github.com/lakshyak789/AI_log_analyzer ↗
LIVE · 12:04 LISTENING
SUGGESTED ANSWER

"I'd start with a heartbeat agent — every minute, ping each service, log latency…"

drop a UI screenshot → get React code back
OSS / 02

A free copilot that sits beside you in the call.

Open-source AI call & interview assistant.

Built as a free alternative to enterprise call assistants — and earned its community traction by refusing to put the useful parts behind a paywall. It listens to live audio for in-call advice, turns UI screenshots into ready-to-use front-end code, and parses your résumé into 15 tailored interview questions.

PythonReactWhisperVision AIWebRTC
github.com/lakshyak789/AI_meeting-copilot ↗
03

Experience

7 roles · 2016 → now

Engineering a Plant Health Management System for Ferroglobe's US operations — a full-stack monitoring platform with admin controls, a Python heartbeat agent, and incident workflows. Offshore, across timezones.

  • Client: Ferroglobe (US) · industrial metals
  • React · FastAPI · PostgreSQL · Python
  • Heartbeat agent pings services every 60s
  • Real-time green/red status dashboard
  • Email-subscription system per service
  • Also: bug fixes & features on a legacy Django REST + React TS app

Designed and built speech-to-text pipelines on Whisper AI — added Tamil support via Whisper.cpp, GPU-pressure queueing, and a custom Confluence diff tool. Full-stack: FastAPI backends, React frontends.

  • Whisper.cpp + FastAPI · FFmpeg chunking (15s)
  • Fine-tuned Whisper on 400h on Kaggle
  • Confluence Page-Diff tool (FastAPI + React)
  • Tampermonkey integration for in-page launch

Worked with the org's web team building React-based pages, then shifted to reporting using Google Apps Script and Looker Studio.

  • React · Apps Script · Looker Studio
  • Internal reporting dashboards

Engineered for Labra.io (US) — front-end UI in TypeScript/React/Redux with Jest, back-end APIs in Go and Python. Owned a ConnectWise CRM integration end-to-end.

  • Client: Labra.io (US) · AWS marketplace SaaS
  • TypeScript · React · Redux · Jest
  • Go · Python · PostgreSQL
  • ConnectWise CRM integration

Carried the front end of a ticketing portal solo. Migrated the marketing site from AngularJS to Gatsby, swapped REST for Apollo GraphQL, shipped a WebRTC video-chat.

  • Autobahn pub/sub · Apollo GraphQL · Gatsby
  • WebRTC video-chat application
  • Jenkins CI · Docker containerization
  • Material-UI · Python APIs

First professional role. Built React UI for a YAML-driven chatbot platform — the kind of system that taught me to read other people's configs carefully.

  • React · YAML flow authoring

Worked on an electronic-health-record back end in CakePHP. The first time I saw what 'legacy' really meant.

  • CakePHP · MySQL · EHR domain
04

Stack

tools & platforms
S/01

Front-of-house

ReactTypeScriptReduxNext.jsGatsbyApollo GraphQLMaterial UIJest
S/02

Back-of-house

PythonFastAPIDjangoNodeGoPostgreSQLRESTGraphQL
S/03

AI & audio

WhisperWhisper.cppFFmpegKaggle (fine-tune)Prompt orchestration
S/04

Glue & ops

AWSDockerLinuxJenkinsWebRTCAutobahn pub/subSlack/Email automationConfluence/Jira scripts
05

Contact

response within 24 hours

Got a system that needs listening to?

Say
hello

Open to contract work, full-time roles, and small open-source collaborations — especially anything sitting between AI, audio, and infrastructure.

Book a call 30 min · cal.com ↗ Email Lakshyak789@gmail.com ↗ GitHub @lakshyak789 ↗ LinkedIn @Lakshya Khanna ↗ Résumé Download CV ↓
Where Dehradun, IN · --:-- IST