Skip to content
Shashank.

Ideas, worked through.

I’m Shashank. I like understanding a problem, working out what the software should do and building a version people can try. My focus is applied AI, product thinking and practical implementation.

How I contribute

Understand. Connect. Build. Check.

Turn a problem into a product flow

I work through who needs to do what, which information they need and what should happen next. That gives an idea a clear shape before adding features.

See the workflow decisions

Connect AI to a useful task

I integrate models with data, tools and a review step. The goal is an answer someone can check and use, with its sources and limits visible.

See an AI workflow

Work through the numbers

I compare options, test assumptions and look at mistakes as well as successful results. My machine-learning demonstration includes a held-out test and the cases the model missed.

Explore the NLP evaluations

Make the solution usable

I connect interface behaviour, application logic and integrations around the task. Client work has included websites, apps, configuration tools and AI automation.

See an interactive application

My working approach

Start with the person using it.

I ask what is confusing or taking too much time, map the important steps and build a focused version. Then I try it, check the awkward cases and refine the experience.

Understand the taskChoose a useful scopeBuild a working versionCheck the behaviourRefine with feedback
More about how I think

Freelance experience

From an idea to a working system.

My client work spans AI agents, sensor analytics, business automation and custom CRM applications. I work across understanding the problem, implementing the solution and making the result usable.

Task-focused AI agents

Built agents for defined business tasks, connecting paid model services with the tools used in the workflow.

Task design · Model APIs · Tool integration

Read the work story

ML sensor analytics

Worked with industrial sensor readings to study operating behaviour and identify anomalies in bottle-cap and automobile-parts manufacturing.

Sensor calibration · Applied ML · Investigation

Read the work story

MSME workflow automation

Built workflow automation for smaller businesses, connecting tools around their day-to-day processes and repetitive manual work.

Process mapping · APIs · Workflow implementation

Read the work story

Organization-specific CRMs

Built focused CRM systems for organizations that wanted software shaped around their work, with reliability and everyday usability guiding the scope.

Requirements · Application logic · User experience

Read the work story

The portfolio also includes independent implementations and public-data evaluations. Those make the technical approach inspectable while the original client details remain private.

Explore the work
Technical toolkit and cloud exposure

Applied AI / ML

  • Machine learning and anomaly detection
  • Supervised NLP and TF-IDF classification
  • Sentence embeddings and hybrid retrieval
  • scikit-learn and ONNX Runtime
  • Model tool calling and LangChain
  • Evaluation, source checks and human review

Applications & data

  • Python, SQL and FastAPI
  • Next.js, React and TypeScript
  • Flutter application development
  • Client AI integrations: OpenAI and Google Gemini

Product & delivery

  • Requirements discovery and workflow design
  • Interactive prototyping and interface design
  • Integration, testing and reviewable handovers
  • Introductory familiarity with AWS and Microsoft Azure.

The project notes show how these tools are used in specific implementations. Cloud services are an area I’m continuing to learn, with introductory AWS and Azure familiarity.