Backend and applied AI engineer

I build production APIs and the AI systems behind them.

Hi, I'm Harshit, a backend engineer in Delhi NCR working with Python, FastAPI, RAG pipelines and voice AI. I like owning a service end to end, from data model to deployment.

Harshit Varshney
OCR API latency
24s → 8s
Production Python
2+ years
Shipped projects
6
// featured work

Case studies

🎙

Voice AI R&D for an AI receptionist

Researched voice cloning and benchmarked speech engines for a real-time conversational product.

DeepgramOmniVoiceHugging FaceSTT / TTSVoice cloning
Read case study

Problem

The platform's AI avatar needed a natural, consistent voice and fast speech handling. Early prototypes on Google Colab had inference too slow for live conversation.

Approach

  • Prototyped voice cloning with open-source models to reproduce a target speaker
  • Benchmarked Deepgram against open-source TTS such as OmniVoice
  • Moved model inference from Colab to Hugging Face

Result

Inference became fast enough for the conversational pipeline, and the team had benchmark data to choose an STT/TTS stack on evidence.

💬

RAG customer-support assistant

An AI chatbot that answers customer queries with grounded, on-brand responses from a live knowledge base.

FastAPIPineconeEmbeddingsLLMsWeb scraping
Read case study

Problem

A brand needed support answers that stayed accurate as product information changed, without hand-maintaining an FAQ.

Approach

  • Built a scraping pipeline to ingest and refresh domain knowledge
  • Embedded content into Pinecone for semantic retrieval
  • Grounded LLM answers in retrieved context through a modular FastAPI service

Result

A low-latency API the frontend could plug into directly, with knowledge that refreshes automatically.

⚡

FoodAI: 3× faster scanning API

Cut a barcode and OCR scanning endpoint from about 24 seconds to about 8 on an AI food-analysis platform.

DjangoPostgreSQLPytesseractAWS S3LLMs
Read case study

Problem

Users scanned a product and waited around 24 seconds for results, far too long for a mobile scanning flow.

Approach

  • Trimmed oversized response payloads
  • Reworked backend processing logic on the hot path
  • Stored scanned images in S3 within the Django REST architecture

Result

Response time dropped to about 8 seconds, roughly 3× faster, alongside personalized health insights built from user profiles.

// more projects

Also built

Applicant tracking system

Candidate workflows, interview scheduling and LLM-powered resume summaries, containerized with Docker and automated email via SMTP.

FastAPIDocker

Multi-tenant RBAC

Tenant-aware access control with data isolation and dynamic role-permission mapping across applications.

FastAPIPostgreSQL

LLM travel assistant

Real-time recommendations and itinerary planning, combining LLMs with live data from SERP APIs.

FastAPIStreamlit
// experience

Where I've worked

Associate Software Engineer · Webmobril Technologies

Feb 2025 – present
  • Design and build backend services and REST APIs powering real-time features, search and scheduling across products.
  • Improved API response times by optimizing PostgreSQL queries and streamlining data flow.
  • Set up CI/CD pipelines with GitHub Actions for faster, more reliable deployments.
PythonREST APIsPostgreSQLGitHub ActionsCI/CD

Software Engineer Intern · Nuage Biztech

Feb 2024 – Sep 2024
  • Built FastAPI backend APIs with authentication and validation, backed by PostgreSQL.
  • Load-tested APIs with Locust to find bottlenecks under concurrent traffic.
  • Wrote unit and integration tests with pytest to reduce regressions.
FastAPIPostgreSQLLocustpytest
// about

A bit about me

I'm a backend engineer based in Delhi NCR, with a B.Tech in Information Technology. I got into backend work because I like the parts of software nobody sees but everybody depends on: the data model, the API contract, the query that has to come back fast.

Currently I'm at Webmobril Technologies, building APIs for real-time features, search and scheduling. Alongside that, I've gone deep on applied AI, from RAG pipelines over vector databases to benchmarking speech engines for a voice assistant. The work I enjoy most is owning a service end to end, from schema to deployment.

Right now I'm writing up what I learned from voice AI R&D, sharpening my problem-solving fundamentals, and looking for a team that's building something real with LLMs.

// toolkit

What I work with

Backend

PythonFastAPIDjangoFlaskREST

AI / LLM

RAGEmbeddingsSemantic searchSTT / TTSHugging Face

Data

PostgreSQLSQLAlchemyPinecone

DevOps

DockerGitHub ActionsAWS S3pytestLocust
// writing

Notes from the work

Voice cloning and STT/TTS: what I tried and what worked

Benchmarks, trade-offs, and why moving off Colab mattered.

Coming soon
// contact

Let's build something.

Always happy to talk about backend and AI work.