You will get Lightning-Fast WebAssembly Applications for Web Performance
Project details
Supercharge your web applications with WebAssembly. Achieve 10-100x performance gains for computational tasks. Perfect for image processing, scientific computing, and data-heavy applications.
Programming Languages
HTML & CSS, JavaScript, C#Coding Expertise
Cross Browser & Device Compatibility, Performance OptimizationWhat's included
| Service Tiers |
Starter
$350
|
Standard
$700
|
Advanced
$1,300
|
|---|---|---|---|
| Delivery Time | 5 days | 9 days | 14 days |
Number of Revisions | 1 | 1 | 1 |
Number of Pages | 1 | 1 | 1 |
Design Customization | - | - | - |
Content Upload | - | - | - |
Responsive Design | - | - | - |
Source Code |
Optional add-ons
You can add these on the next page.
Fast Delivery
+$150 - $250
Additional Revision
+$200
1 review
(1)
(0)
(0)
(0)
(0)
This project doesn't have any reviews.
TG
Tom G.
Apr 5, 2021
Need JS help figuring out wavesurfer.js overlap logic
Stop looking, this is your guy right here, after working with countless of other developers who couldn't figure out the problems me and my team were having, this guy solved it right here.
He's very communicative, fast, and accurate. Plus he overdelivered which is highly appreciated!
He's very communicative, fast, and accurate. Plus he overdelivered which is highly appreciated!
About Smithangshu
Senior ML Engineer | LLM Fine-Tuning, CV, DICOM, IoT | 12 Yrs
Alipurduar, India - 9:05 pm local time
That's what I specialize in — building systems that are engineered precisely for the task at hand, whether that's a fine-tuned model that classifies at scale, a computer vision pipeline that actually works in your lighting conditions, or a DICOM workflow that fits how your clinical team actually operates.
A recent example:
A client needed to classify thousands of business news articles per day into accurate domain categories. General-purpose LLM prompting kept producing inconsistent labels — the kind of output variance that breaks downstream pipelines. I fine-tuned a Qwen2.5 1.5B model specifically on their data and taxonomy. The result was a model that runs entirely on a CPU-only VPS, processes several thousand articles daily, and produces consistent, deterministic labels on every inference — no prompt engineering, no surprises, no cloud API dependency.
A 1.5B model that does one job reliably is worth more in production than a 70B model you're fighting with prompts.
What I work on:
LLM Fine-Tuning — Fine-tuning small language models (1B–13B) for specific, repeatable tasks: classification, structured extraction, domain Q&A. If you need consistent output and want the model running on your own infrastructure without ongoing API costs, this is the approach.
Computer Vision — Systems built around your actual environment. YOLO-based detection, camera calibration, optical flow, real-time analysis, and custom-trained models when off-the-shelf weights don't cut it. OpenCV, PyTorch, custom pipelines.
DICOM and Medical Imaging — PACS workflows, Store SCP/SCU, imaging pipeline automation. Delivered a system that cut a clinical team's preparation time by 75%.
IoT Automation — Biometric systems, industrial monitoring, kiosk infrastructure. Built systems handling thousands of daily transactions in production.
High-Performance Web — WebAssembly, GPU computing, real-time browser applications where standard web tech hits a ceiling.
Code Protection and Licensing — Anti-piracy systems, GDPR/HIPAA-compliant software delivery.
Clients include healthcare institutions, government ministries, aviation companies, and enterprise clients across the US, Europe, Middle East, and Asia.
I have 18+ verified 5-star reviews across 8 years of independent consulting work on other platforms.
If you have a specific problem and want to talk through whether it's something I can help with, message me with the details.
Steps for completing your project
After purchasing the project, send requirements so Smithangshu can start the project.
Delivery time starts when Smithangshu receives requirements from you.
Smithangshu works on your project following the steps below.
Revisions may occur after the delivery date.
Requirement analysis
Core development