Govind Pradeep
contact me

$ whoami

Govind Pradeep

Builder. Automator.

senior automation & ai systems engineer · remote-friendly, works with teams anywhere

$ cat intro.txt

I build systems that act on their own — browsers that navigate and decide, computer-vision models that catch what a human would miss, pipelines that turn manual, repetitive work into something that runs itself. ~7 years focused on automation and applied AI: designing autonomous agents, training and shipping models, and building the infrastructure that lets them run unsupervised.

my work download resume

$ tail -f systems.log --summary

By the numbers

# what the last seven years add up to

[ok] 0 projects shipped — AsimovX & Accubits
[ok] 0 professional AI & automation engineering
[ok] 0 APIs & endpoints built across those systems
[ok] 0 rows of data scraped & processed
[ok] 0 browser extensions built end to end
[ok] 0 AWS Lambda functions in one production pipeline
[ok] 0 trained models shipped to on-device iOS

$ ls -la capabilities/

What I build with

# seven areas, and what each one is actually built on

AI & LLM Integration AWS Bedrock · Nova Pro · Nemotron · Hugging Face · Fine-tuned LLaMA · RAG · Rasa · DialogFlow Computer Vision & Machine Learning Keras · TensorFlow · ONNX Runtime · CoreML · Object detection · Model benchmarking Team Leadership & QA Tooling Team leadership · QA workflow design · Internal tooling · Annotator attribution Cloud & Backend Infrastructure AWS Lambda · DynamoDB · S3 · ECR · EC2 · Flask · Express · Docker Data Engineering at Scale ETL · Deduplication & merge logic · Structured extraction · Atomic writes Browser Automation & Scraping Playwright · Selenium · Chrome Extensions · CDP control · Anti-bot resilience · Session portability Mobile & On-Device AI SwiftUI · Cognito Identity · On-device inference

what each one means in practice ->

$ ps aux --systems

Systems I've built

# thirteen systems, live or run to completion — not demos

status system pipeline metric
[run] Marketplace Negotiation Agent capture -> price -> negotiate negotiates in parallel, unattended
[ok] Sports Capture Quality Control upload -> checks -> AI fallback -> video 86% recall · ~16s render
[ok] Launch Monitor ML Suite collect -> engineer -> train -> ship to iOS 20.5 -> 1.99 km/h error
[ok] Tournament Data Agent discover -> extract -> read flyers -> dedupe 178 tournaments extracted
[ok] Professional Network Automator browse -> engage -> alert 1,541 items · 98.8% success
[ok] Web Scraping Fleet extension -> local server -> dataset 22,000+ records · 5 platforms

$ cat archive/accubits-systems.log

# shipped 2019–2023 at accubits technologies, for real clients

status system pipeline metric
[ok] Deep Listening ingest -> extract -> LLM -> retrieve text, image, audio and video
[ok] CipBot WhatsApp -> chat flow -> database Rasa · live for a pharma team
[ok] Trajectus IVR call -> voice IVR -> query handling voice-first, not text
[ok] Coglide — Cori & Gx email -> extract -> sync -> database where the scraping work began
[ok] Emotyx live CCTV -> backend -> insights real-time video, not batch
[ok] Attendance Management System records -> backend -> image processing full backend ownership
[ok] Other Projects — the long tail, same instinct

read the full write-ups ->

$ cat experience.log

Where I've been building

  1. Most recent

    Senior Automation & AI Engineer

    Feb 2024 to Present
    • Own the production AWS pipeline behind a sports-technology product's shot data — nine Lambda functions spanning ingest, quality control, alignment, scoring, daily reporting and alerting.
    • Built the automated shot quality-control system: parameter checks, then CNN/ONNX visual verification with an AI fallback, plus a rendered annotated video for human review. 86% recall on range-mode ball-impact detection.
    • Lead a team of annotators producing ground-truth data, and built the desktop tooling they use daily — per-person attribution, a correction and verification workflow, and live physics feedback.
    • Trained and shipped three neural networks predicting spin rate, spin axis and club speed on ~2.4M rows, cutting club-speed error from 20.5 km/h to 1.99 km/h.
    • Built autonomous agents that negotiate, extract and scrape unattended, plus a fleet of five production scrapers collecting 22,000+ records.
    • AWS Bedrock
    • Playwright
    • Lambda
    • DynamoDB
    • ONNX
    • CoreML
    • Flask
    • SwiftUI

  2. AI Engineer

    Aug 2019 to Dec 2023
    • Deep Listening — a multi-modal extraction platform: object detection, tagging and description on images; translation and transcription on audio; summarization and keyword extraction on text; a fine-tuned LLaMA with vector-database retrieval.
    • CipBot — a Rasa-built WhatsApp chatbot connecting a pharma company's procurement team with vendors, with role-based flows and an admin panel.
    • Trajectus (Computer Instruments) — a DialogFlow IVR chatbot handling medical query calls.
    • Coglide — the "Cori" email service extracting contact details from inbound mail, a "Gx" browser extension syncing marketing data, and the scraping and automation behind both.
    • Emotyx — a backend extracting actionable customer insights from real-time CCTV footage.
    • Also delivered a GPT-3.5 meeting-minutes generator, SQL-injection-hardened API verification tooling, a Selenium scraper across social networks, and a PPE-kit detection module.
    • Hugging Face
    • Fine-tuned LLaMA
    • Rasa
    • DialogFlow
    • Python
    • Selenium
    • GPT-3.5

  3. B.Tech, Computer Science

    Sree Chitra Thirunal College of Engineering

    2019 Thiruvananthapuram
    • Final year project — a supply chain system built on Hyperledger Sawtooth, a permissioned blockchain framework.
    • Higher Secondary, Computer Science — CBSE, Kendriya Vidyalaya Pattom, 2015.
    • Certification: Python: Programming Efficiently — LinkedIn Learning.

$ cat about-me.txt

About me

Maybe I'm lazy — but that's exactly what made me automate everything. Most of what's on this page started as "this manual work is stupid, I could automate it" and turned into something that runs on its own. That's the part I enjoy: the moment a system stops needing me.

Based in Bangalore, India, working with teams anywhere.

Off the clock I read a lot, and I'm a fairly serious gamer. Mostly Dota 2, where I play as devilsnare — currently rank 4300 on the SEA leaderboards, which takes about as much stubbornness as debugging someone else's Selenium script at 2am. Same instinct, different arena: learn the system, find where it breaks, get better at it.

$ ls blog/

Writing

[post] How I Created My First Discord Bot Using JavaScript

A walkthrough of building a Discord bot from scratch in JavaScript — setting it up, wiring the event handlers, and getting it running.

Read on Medium →

$ whois govind

Get in touch

Available for new work

Bangalore, India — remote-friendly · local time --:--

Let's talk for 30m Pick a slot that works for you — no forms, no back-and-forth. Book a call Send an email Longer brief, a role, or something you want built. I read everything. govindpradeep1998@gmail.com

$ cat links/