Sankalp's Portfolio Assistant

Online · trained on resume & projects

Things I've built & shipped

Production systems, applied research, and the kind of work that ships on Friday and stays up on Monday. Tap any card to flip →

Internships

Publication

3.81

GPA

Experience

🚀

Feb 2026 → Present Now

Stealth Startup

Software Developer Intern · AI / Backend · Remote

Built 9 personalized tool-calling agents and a pgvector semantic classifier across ~900 O*NET occupations. Shipped 20+ REST APIs, a React Native app on TestFlight, and LLM eval harnesses to catch prompt regressions.

AI Agents GCP pgvector React Native

Flip

What I'm doingStealth Startup · AI / Backend

Built and shipped 9 personalized tool-calling agents (onboarding, matching, classification) that adapt behavior dynamically to each user's profile and work history, backed by structured extraction pipelines with system prompts and few-shot examples.
Built pgvector-based semantic matching classifying free-form profile text against ~900 O*NET occupations — 87% classification consistency, eliminating manual occupation tagging from the onboarding workflow.
Designed and shipped 20+ REST APIs (14 internal services), deployed the full backend on GCP, and shipped a React Native mobile app with phone OTP auth — currently in TestFlight.
Built evaluation harnesses and regression tests for LLM outputs to catch prompt regressions across agent iterations, establishing repeatable quality baselines for AI-driven features.

⚙️

Jul → Sep 2025 • Chicago, IL

Qyrus

AI Engineer Intern

Built a production RAG system indexing 220K+ embeddings. Improved precision@5 by 22% via CLIP + Cohere ReRank v3. Fault-tolerant ingestion at 20K+ items/hour on AWS.

RAG AWS Pinecone FastAPI

Flip

What I shippedQyrus RAG Platform

Built and deployed a production RAG system (FastAPI, Redis, Pinecone, Docker) indexing 220K+ embeddings for low-latency search.
Designed embedding + reranking workflows with CLIP and Cohere ReRank v3 — +22% precision@5 while balancing latency and cost.
Architected a fault-tolerant async pipeline on AWS (S3, SQS, async workers, batch upserts) with backpressure and retry logic — 20K+ items/hour.
Sustained 300+ vectors/min ingestion throughput with structured metrics and distributed logs.
Integrated LLM generation with grounding, prompt templates, and fallback logic; validated end-to-end with PyTest + Postman.

🛠️

Jun 2023 → Jul 2024 • Bangalore, IN

Pcloudy

Software Engineering Intern

Shipped Qpilot.AI MVP — automated test generation across 1,200+ OEM configs with 78% execution success and 90% reduction in test time. Kafka log pipeline at 33K+ events/min.

Flask Kafka AWS Docker

Flip

What I shippedQpilot.AI MVP

Designed and shipped the Qpilot.AI MVP (Flask, Python) — automated script generation and self-healing execution across 1,200+ OEM device configs.
Achieved 78% execution success and reduced end-to-end test time by 90%, significantly lowering manual intervention.
Built Python/Flask microservices on AWS with Kafka-based streaming log pipelines for real-time failure triage at 33K+ events/min.
Automated build, test, deploy via GitHub Actions; containerized with Docker for iterative rollouts.
Validated services with Postman + PyTest; added logging and monitoring for production reliability.

Projects

🧠

2024 → Present Active

Sage

Local-First Personal AI Agent

Personal Agent that runs entirely on-device via Ollama (Llama 3.1, Mistral). Orchestrator Agent executes a Thought → Action → Observation loop across 8 LLM-callable tools — CLI, WhatsApp, and Web UI on one FastAPI backend. Fully private, no external API calls.

Ollama ChromaDB FastAPI Docker

Flip

Project deep-diveSage · Local-First Personal AI Agent

Built a full-stack agentic AI assistant (10K+ lines, 60+ files) running entirely on-device via Ollama (Llama 3.1, Mistral) with ChromaDB for semantic retrieval, SQLite for persistent memory, and 8 LLM-callable tools including web search, Gmail triage, and document Q&A.
Designed an Orchestrator Agent that executes a Thought → Action → Observation loop to autonomously decompose and complete multi-step goals across specialized sub-agents with no hardcoded step routing.
Shipped multi-channel access (CLI, WhatsApp via Twilio, Web UI) on the same FastAPI backend with selective RAG, proactive scheduling, and Docker deployment.
Fully private — no external LLM API calls; all models run on-device via Ollama.

GitHub

🔍

Apr 2025 → Present Pilot

SafeBot

Investigation platform for Riverside PD

Human-in-the-loop email triage and suspect-flagging with hybrid keyword rules + LLM risk scoring. Google OAuth, RBAC, audit logs. Deployed on GCP Cloud Run with CI/CD.

GCP LLM React PostgreSQL

Flip

Project deep-diveSafeBot · Applied AI for Investigations

Built a production investigation platform for Riverside Police Department processing 10K+ emails/hour for triage and suspect flagging.
Hybrid keyword rules + LLM-based risk scoring with RBAC, escalation policies, and review gates aligned to legal requirements.
Shipped REST APIs with PostgreSQL schema migrations and React + TypeScript dashboards with Google OAuth, admin console, and audit logs.
Deployed on GCP Cloud Run with Redis caching and autoscaling via GitHub Actions CI/CD.
Currently in pilot discussions for deployment to 5 RPD investigators.

📊

2023

Test Device Recommendation Engine

ML-driven device matrix selection

Clustered 1K+ OEM configurations and ranked similarity over 125K+ log lines via TF-IDF. Replaced manual heuristics with data-driven scoring and cut device selection time to under 90 seconds.

ML TF-IDF Clustering

Flip

Project deep-diveTest Device Recommendation Engine

Built a recommendation system to automate test-device selection for QA teams managing massive device matrices.
Clustered 1K+ OEM configurations and analyzed 125K+ log lines using TF-IDF and similarity-based ranking.
Replaced manual heuristics with data-driven scoring — device selection time cut to under 90 seconds.
Integrated as a recommendation layer into the broader Pcloudy test automation platform.

📱

2023 • Published 2024

Android Log Search Utility

Real-time log parsing & classification

Parsed 250K+ Android log lines with failure classification. Published in the International Journal of Applied Engineering and Technology, London, 2024.

Node.js MongoDB Published

Flip

Project deep-diveAndroid Log Search Utility

Built a real-time Android log parsing and search system (Node.js, MongoDB) processing 250K+ log lines.
Added failure classification and recommendations for log lines, surfacing error patterns from noisy logs.
Improved debugging efficiency for engineers working across high-volume device fleets.
Work published as "Application Specific Log Parser" in the International Journal of Applied Engineering and Technology, London, 2024.

Read Paper

🔎

2025 • CS242 @ UCR

Fandom Wiki Search Engine

Information retrieval across Fandom wikis

Indexes and retrieves content across Fandom wikis — characters, lore, game mechanics, community knowledge — using keyword retrieval, ranking algorithms, and NLP enhancements.

IR NLP Python

Flip

Project deep-diveFandom Wiki Search Engine

Built a full information retrieval pipeline for searching across Fandom wikis — characters, lore, game mechanics, and community knowledge.
Implemented inverted index construction, TF-IDF scoring, and BM25 ranking for relevance-based retrieval.
Added NLP enhancements including query expansion and stemming to improve recall.
Final project for CS242: Information Retrieval and Web Search at UC Riverside.

GitHub

Publications

2024 • London, UK • Peer Reviewed

Application Specific Log Parser

International Journal of Applied Engineering and Technology

Real-time Android log parsing and failure classification system processing 250K+ log lines — surfacing error patterns from noisy device logs and improving debugging efficiency for high-volume QA fleets.

Node.js MongoDB Log Analysis Android

Read Paper

Education

🎓

Sept 2024 → Mar 2026Graduated

University of California, Riverside

MS in Computer Science

GPA 3.81 / 4.0

🎓

Dec 2020 → May 2024Graduated

PES University, Bangalore

BTech CS · Machine Intelligence & Data Science

GPA 8.3 / 10

Let's build something.

Open to SWE / AI Engineer roles · Riverside, CA

Email LinkedIn GitHub

Places I've wandered

Eight countries, three continents. Tap a glowing country to see photos.

Countries

Continents

Loading map…