Ahmedabad, India · Open to opportunities

Pratham Shah.
AI Engineer

I build products where full-stack engineering meets applied AI — LLM-powered agents, production APIs, and data systems that ship real, measurable impact.

View my work Contact me

0+

Production API endpoints

0+

Repositories tracked by my tooling

0%

Manual reporting effort reduced

01 · About me

An engineer who ships across the stack — and into the model.

I'm a software engineer with a BE in Artificial Intelligence & Machine Learning and production experience at an enterprise SaaS platform, where I design and deploy LLM-powered analytics workflows, REST APIs, and role-based dashboards used across 50+ tracked repositories.

My range covers the whole product: React / Next.js frontends, Django / FastAPI / Node.js backends, PostgreSQL data layers, and the applied-AI layer on top — RAG pipelines, agent tool loops with LangChain / LangGraph, and classical ML with XGBoost and SHAP, validated with real statistics.

Most recently I built Aster, a multi-user AI personal assistant that connects Gmail and WhatsApp, generates scheduled daily briefings, and chats through a streaming agent with live tool access.

Currently

Junior Software Engineer @ Akkomplish Group

LLM analytics · Django REST · dashboards

Focus

Applied AI & full-stack product engineering

Agents · RAG · APIs · data science

Beyond the code

Statistical rigor meets product impact

A/B testing · experimentation · measurable outcomes

02 · Skills

Tools I think in.

03 · Experience

Where I've shipped.

Junior Software Engineer

Jun 2025 — Present

Akkomplish Group · Ahmedabad

  • Designed and deployed LLM-powered analytics workflows (OpenAI, Ollama) for automated code-review analysis and progress summarization in the BAS TrustDesk platform.
  • Engineered 25+ production REST endpoints with Django REST Framework, cutting average response time ~20% via query optimization and efficient serialization.
  • Built RAG-style prompt pipelines and context-aware summarization, reducing manual reporting effort ~30% across 50+ tracked repositories.
  • Delivered role-based dashboards surfacing project-health KPIs and code-quality metrics for stakeholders; collaborated on model evaluation and prompt iteration in an agile team.

Software Developer Intern

Jan 2025 — May 2025

Elixir Techne · Ahmedabad

  • Built data-driven API workflows and reusable Python service modules powering analytics and reporting features.
  • Implemented data validation, transformation, and integration across multiple sources for reliable application inputs.
  • Optimized request handling and data flow, improving end-to-end application performance and throughput.

04 · Projects

Selected work.

Featured · Flagship build

Next.js 16 React 19 PostgreSQL Trigger.dev MCP

Aster — AI Personal Assistant

A multi-user AI assistant that connects Gmail and WhatsApp, turns your day into an AI-generated brief, and lets you chat with a streaming agent that pulls live data from your accounts. Namespaced MCP tool servers per provider, a bounded agent tool loop over an OpenAI-compatible API, pairing-code WhatsApp auth via Baileys, cookie sessions on Postgres, and scheduled briefings fanned out through an atomic Trigger.dev cron pipeline.

  • Streaming NDJSON agent chat with live tool-status events
  • Per-user OAuth tokens — multi-user by design, no shared state
  • AI daily briefs from a pluggable provider registry
  • Idempotent scheduled briefing runs, claimed atomically in SQL
View on GitHub Live demo — coming soon

05 · Education

Foundation.

LJ University

2021 — 2025

B.E. — Artificial Intelligence & Machine Learning

Coursework: Machine Learning, Neural Networks, Probability & Statistics, Data Structures & Algorithms, DBMS, Software Engineering, Web Technologies.

07 · Achievements

Numbers that shipped.

0+

Production REST endpoints designed, built, and deployed at Akkomplish.

0%

Reduction in manual reporting effort from my RAG summarization pipelines.

0.0

ROC-AUC on churn prediction with tuned XGBoost — at 80% churn-class recall.

0K

Users in the A/B test I analyzed end-to-end — bootstrap, power, effect size.

08 · Contact

Let's build something intelligent together.

I'm open to software engineering, applied AI, and data roles or a conversation about an idea you want shipped. Reach out any way you like.

Email me Download Resume
prathamshah8487@gmail.com +91 84870 17317 github.com/Pratham8487 linkedin.com/in/pratham1212