Skip to content
View TalhaSikandar's full-sized avatar
🙃
Hardly working.
🙃
Hardly working.

Block or report TalhaSikandar

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
TalhaSikandar/README.md
╔══════════════════════════════════════════════════════════════════╗
║                                                                  ║
║                 T A L H A   //   C O N T R O L                   ║
║                                                                  ║
║        AI SYSTEMS · SOFTWARE ENGINEERING · REAL-WORLD OPS         ║
║                                                                  ║
╚══════════════════════════════════════════════════════════════════╝

Talha Sikandar

AI/ML Engineer · Software Engineer · Systems Builder

I build intelligent systems that move beyond notebooks— into products, infrastructure, and real-world operations.


Portfolio LinkedIn Email


01 // SYSTEM OVERVIEW

operator:
  name: Talha Sikandar
  role: AI/ML Engineer and Software Engineer
  base: Lahore, Pakistan
  education: BS Computer Science — Information Technology University
  specialization:
    - Industrial AI
    - Machine Learning Systems
    - Full-Stack Engineering
    - MLOps and Model Deployment
    - Linux and Developer Infrastructure

system_state:
  status: ONLINE
  mode: BUILDING
  availability: OPEN_TO_HIGH_IMPACT_WORK

primary_directive:
  "Convert complex operational problems into reliable,
   intelligent and production-ready systems."

CURRENT OPERATIONS

● Designing production AI systems
● Building end-to-end ML pipelines
● Developing intelligent applications
● Experimenting with systems and infrastructure
● Turning research into usable products

ENGINEERING PRINCIPLES

01. Understand the real problem
02. Build the smallest valid system
03. Measure before optimizing
04. Design for failure
05. Deploy, observe and improve

02 // LIVE CAPABILITY MATRIX

Machine Learning       █████████████████░░░  Production Systems
Python Engineering     ██████████████████░░  Primary Language
Full-Stack Systems     ████████████████░░░░  Product Development
Linux & Infrastructure ████████████████░░░░  Daily Environment
MLOps & Deployment     ███████████████░░░░░  Active Development
Systems Programming    ██████████████░░░░░░  Continuous Learning

I am most interested in work where machine intelligence, software engineering, and real-world operations intersect.


03 // MISSION ARCHIVE

Mission File PX-01

STATUS

FIELD DEPLOYED

DOMAIN

INDUSTRIAL AI

CLEARANCE

FLAGSHIP

PlungerXcel

AI-driven closed-loop optimization for plunger-lifted oil and gas wells.

Traditional well controllers operate using fixed timers and manually configured setpoints. PlungerXcel transforms each production cycle into feedback for the next one.

Mission objectives

  • Process live well telemetry and cycle data
  • Evaluate pressure, flow and arrival behavior
  • Recommend optimized shut-in and afterflow settings
  • Apply operational constraints and safety limits
  • Continuously learn from the response of each well

Operational result

The system has been tested across multiple real production wells and was designed to connect telemetry, machine intelligence and field control in one continuous optimization loop.

Technology signal

Python Machine Learning Time Series Optimization FastAPI React Go Kafka InfluxDB AWS

View the product system →


Mission File PM-02

STATUS

PROTOTYPE ACTIVE

DOMAIN

PREDICTIVE AI

TARGET

ROTATING ASSETS

Predictive Maintenance Intelligence

An AI-assisted monitoring and diagnostic system for industrial pumps and mixers.

The system is designed to move beyond basic dashboards by connecting condition monitoring with fault interpretation and maintenance guidance.

Mission objectives

  • Analyze vibration, RPM and temperature telemetry
  • Identify abnormal operating patterns
  • Detect likely mechanical fault categories
  • Estimate machine health and failure risk
  • Present actionable maintenance recommendations

Monitored failure modes

Bearing Faults · Misalignment · Unbalance Mechanical Looseness · Cavitation · Lubrication Issues

Engineering references

ISO 20816 API 610 API 670 FFT Analysis Anomaly Detection ThingsBoard


Mission File ML-03

STATUS

RESEARCH COMPLETE

DOMAIN

DEEP LEARNING

BUILD TYPE

FROM SCRATCH

MegaMan-CO

A custom Long Short-Term Memory network implemented in pure Python for next-day stock-price prediction.

Rather than relying entirely on high-level deep-learning abstractions, this project explores the internal mechanics of recurrent neural networks.

Systems investigated

  • Recurrent state propagation
  • LSTM gates and memory cells
  • Forward and backward propagation
  • Sequence preprocessing
  • Time-series prediction
  • Model evaluation

Inspect the mission repository →


Mission File SYS-04

STATUS

ALWAYS EVOLVING

DOMAIN

LINUX SYSTEMS

ENVIRONMENT

PERSONAL LAB

Developer Environment and Dotfiles

A version-controlled Linux development environment built for speed, consistency and experimentation.

The repository contains personal system configurations used to reproduce and manage development environments across Linux systems.

Mission objectives

  • Automate workstation configuration
  • Preserve reproducible development environments
  • Improve terminal-centered workflows
  • Experiment with Linux tools and window management
  • Reduce setup time across machines

Enter the configuration repository →


04 // OPERATOR MODEL CARD

┌──────────────────────────────────────────────────────────────────┐
│ MODEL NAME        TalhaSikandar                                  │
│ MODEL FAMILY      Engineer / Researcher / Product Builder         │
│ VERSION           2026.07                                        │
│ DEPLOYMENT        Lahore, Pakistan                                │
│ CURRENT STATE     Actively learning and building                  │
└──────────────────────────────────────────────────────────────────┘

Model description

TalhaSikandar is a production-oriented engineering model trained to work across the complete lifecycle of an intelligent system:

flowchart LR
    A["Operational Problem"] --> B["Research"]
    B --> C["Data Pipeline"]
    C --> D["AI / ML Model"]
    D --> E["Application"]
    E --> F["Deployment"]
    F --> G["Observation"]
    G --> B
Loading

Intended use

+ Building production-grade AI and ML systems
+ Developing full-stack intelligent applications
+ Transforming ambiguous ideas into working products
+ Designing industrial monitoring and optimization systems
+ Creating reliable data and inference pipelines
+ Solving unfamiliar technical problems through research

Training distribution

Knowledge domain Approximate allocation
Building and debugging 35%
AI and machine learning 25%
Software and system design 20%
Research and experimentation 15%
Renaming variables repeatedly 5%

Known behaviors

[+] Questions assumptions before implementing
[+] Enjoys understanding systems below the abstraction layer
[+] Prefers working prototypes over endless presentations
[+] Treats deployment as part of engineering—not an afterthought
[+] May attempt to automate a task after performing it twice

Current limitations

[-] Can spend too long perfecting development environments
[-] Occasionally turns a small experiment into a complete platform
[-] Frequently opens one more terminal than strictly necessary
[-] Performance may correlate with available coffee

05 // TECHNOLOGY MODULES

Intelligence Layer

Python PyTorch TensorFlow scikit-learn Pandas NumPy

Application Layer

React JavaScript FastAPI Node.js Tailwind CSS

Systems Layer

Linux C C++ Go Docker Git

Data and Infrastructure

PostgreSQL InfluxDB Apache Kafka AWS GitHub Actions


06 // ENGINEERING PIPELINE

                           TALHA ENGINEERING LOOP

     ┌───────────────┐
     │   DISCOVER    │
     │ Find the real │
     │    problem    │
     └───────┬───────┘
             │
             ▼
     ┌───────────────┐
     │   RESEARCH    │
     │ Data, systems │
     │  constraints  │
     └───────┬───────┘
             │
             ▼
     ┌───────────────┐
     │   PROTOTYPE   │
     │ Test the core │
     │  assumption   │
     └───────┬───────┘
             │
             ▼
     ┌───────────────┐
     │    ENGINEER   │
     │ Build reliable│
     │   software    │
     └───────┬───────┘
             │
             ▼
     ┌───────────────┐
     │    DEPLOY     │
     │ Connect model │
     │ and operation │
     └───────┬───────┘
             │
             ▼
     ┌───────────────┐
     │    OBSERVE    │
     │ Measure, learn│
     │  and improve  │
     └───────┴───────┘

07 // CURRENT RESEARCH QUEUE

current_focus = {
    "industrial_ai": [
        "closed-loop optimization",
        "predictive maintenance",
        "time-series intelligence",
    ],
    "machine_learning": [
        "model evaluation",
        "adaptive systems",
        "real-time inference",
    ],
    "software_engineering": [
        "distributed systems",
        "production architecture",
        "developer infrastructure",
    ],
    "exploration": [
        "AI agents",
        "GPU orchestration",
        "low-level systems",
    ],
}

08 // TELEMETRY

Talha's GitHub statistics Talha's most-used languages
Talha's contribution streak

GitHub language statistics describe public repository composition—not the complete range or depth of my engineering experience.


09 // CONNECTION PROTOCOL

$ ./find-collaborator \
    --skills "AI/ML, software engineering, industrial systems" \
    --mindset "curious, practical, impact-driven"

Searching...

MATCH FOUND
Name: Talha Sikandar
Status: Available for meaningful engineering challenges

I am interested in opportunities involving:

  • Production AI and machine-learning engineering
  • Industrial automation and intelligent monitoring
  • Full-stack AI products
  • MLOps, infrastructure and model deployment
  • Research-oriented software engineering
  • Challenging systems that require learning beyond familiar tools

Have a difficult problem worth solving?

Start a Conversation



SYSTEM MESSAGE:
The best projects begin with an interesting problem.

Designed as an engineering control room—not a conventional résumé.

Pinned Loading

  1. megaman-co megaman-co Public

    A fully custom LSTM-based next-day stock price predictor written from scratch in Python (using numpy) — no PyTorch, no TensorFlow. Built as an academic project to understand LSTM internals, backpro…

    Jupyter Notebook 2 1

  2. bluefin_hr bluefin_hr Public

    A lightweight C framework for hot-reloading shared libraries at runtime using dynamic linking. Designed as an academic project, it enables seamless updates to .so files without restarting the appli…

    C 1

  3. fourier_t fourier_t Public

    A C-based academic project demonstrating signal processing using Discrete Fourier Transform (DFT) and Fast Fourier Transform (FFT). Includes custom implementations for both complex and sine/cosine …

    C 1

  4. dotfiles dotfiles Public

    A minimal, stow-powered dotfiles setup for managing and version-controlling personal Linux development environment configurations. Portable, clean, and ideal for automation or fresh setups.

    Lua 1

  5. krypton-f krypton-f Public

    Krypton is an academic full-stack web application built to simplify supply chain management. Developed using Django (REST API) and Vue.js, it enables seamless company, product, inventory, warehouse…

    Python 1

  6. fat_fs fat_fs Public

    This project is a custom implementation of the FAT (File Allocation Table) file system written in C. It simulates a virtual disk environment. The system manages file metadata, handles cluster alloc…

    C 1