Open to new opportunities · San Francisco Bay Area
About Me
Applied AI Engineer · RAG, Agents & AI Evaluation
Applied AI Engineer with an M.S. in Statistics who builds retrieval and agent systems and measures — with statistics, not vibes — whether they actually got better.
Ran a five-line 3D-printing and laser-engraving business in India for five years, hit a data problem that outgrew spreadsheets, and taught himself Python to solve it — then decided that was the more interesting problem, moved to the US in 2023 to study statistics, and has been building production AI systems since.
At OpGov.ai, an initiative of the non-profit Lead for Earth, I took over as sole technical owner of the platform after the original lead developer left, and now run six production systems end to end: an AI summarization service, a transcript/ASR pipeline, a hybrid RAG and agent platform (ASK NOHA, orchestrated with LangGraph and exposed as governed tools through an MCP server with human-in-the-loop approval on sensitive actions), a publishing platform with role-based permissions, a multi-channel campaign communications platform used in a real election, and a geospatial field operations dashboard used in a real campaign. I also built a canonical public-record data platform spanning 1.7M+ profiles, and I administer everything myself — the public site on Vercel, one service on Azure Container Apps, and the rest, including ASK NOHA's agent runtime and MCP server, on a Linux server I run end to end.
Before OpGov, I advised a pre-funding hardware founder on an edge-AI pet wearable — turning a concept with no spec into a working Raspberry Pi prototype running on-device YOLO object detection. Below the API layer, I also work on GPU memory, quantization and local inference performance.
I hold a Master's in Statistics (Data Science) from CSU East Bay (GPA 3.9, Heebok Park Scholar) and a B.E. in Mechanical Engineering from University of Mumbai. Winner of 3rd Prize at the MCP × A2A Hackathon for a multimodal autonomous agent system.
Core Skills
AI / Retrieval
RAG
KG-RAG
Hybrid Retrieval
BM25
Dense Retrieval
Reciprocal Rank Fusion
CrossEncoder Reranking
Knowledge Graphs
Neo4j
Entity Resolution
Weaviate
FAISS
Prompt Engineering
Agent Orchestration & Governance
LangGraph
Model Context Protocol (MCP)
Agent Orchestration
Tool Use
Human-in-the-Loop Approval
Policy Enforcement
Autonomy Boundaries
Multi-Agent Systems
Evaluation & Reliability
LLM Evaluation
Golden Datasets
Hard Negatives
Recall@K
MRR
Bootstrap Confidence Intervals
Ablation Studies
LLM-as-Judge
Cohen's Kappa
Guardrails
Prompt-Injection Defense
OpenTelemetry
Langfuse
Statistical Inference
Systems & Engineering
Python
FastAPI & Flask
REST APIs
Pydantic
Docker & CI/CD
Azure Container Apps
Linux Server Administration
React
Next.js
TypeScript
Data & ML Systems
MongoDB
PostgreSQL
ETL
Identity Resolution
Geospatial (GeoJSON)
PyTorch
Quantization
LoRA / QLoRA / PEFT
CUDA
GPU Profiling
Inference Benchmarking
Computer Vision & Edge AI
Computer Vision
Object Detection
YOLO
On-Device Inference
Edge AI
TensorFlow Lite
Raspberry Pi
Pi AI Camera
Multimodal Sensing
Hardware & Product Design
CAD
Mechanical Design
3D Printing
Rapid Prototyping
Design for Manufacture
Hardware/Software System Architecture
Languages
English· Professional
Hindi· Native
Marathi· Native
Gujarati· Native
Currently Exploring
Production LLM systems
Agent orchestration & governance
AI evaluation with real statistics
Retrieval quality & RAG
Observability
Local LLM inference
GPU memory & quantization
Scalable backend design
Experience & Education
WorkStartup / FoundingEducation
1
Founding AI Engineer
OpGov.ai · an initiative of Lead for Earth (non-profit)
Aug 2025 – Present
- ▸Took over as technical owner of the entire platform after the lead developer departed, growing ASK NOHA into an agentic system orchestrated with LangGraph and exposed as governed tools through an MCP server, with human-in-the-loop approval on sensitive actions and a policy layer enforcing autonomy boundaries
- ▸Developed a versioned evaluation harness with bootstrap confidence intervals on retrieval deltas, ablation studies, and an LLM-as-judge calibrated against human labels via Cohen's kappa — used as a release gate, not a report
- ▸Shipped four more production systems end to end: a publishing platform with a role-based permissions model I designed, a multi-channel campaign communications platform used in a real election, and a geospatial field operations dashboard used in a real campaign
- ▸Engineered a canonical MongoDB platform unifying 1.7M+ public-record profiles and ~4M historical participation records, reconciling a 734,457-record snapshot into 20,060 new profiles while isolating a single ambiguous identity for review
- ▸Administer the platform's infrastructure end to end — the public site on Vercel, the summarization service on Azure Container Apps, and everything else, including ASK NOHA's agent runtime and MCP server, on a Linux server I run myself
- ▸Own product design, system architecture, implementation, deployment and operations across the platform, coordinating a team of up to four including a remote developer
Python
LangGraph
MCP
Next.js
MongoDB
Neo4j
Weaviate
FAISS
Gemini
Azure
Linux
2
AI Engineer
OpGov.ai
Jan 2025 – Aug 2025
- ▸Joined as the platform's AI engineer and built its first AI service: the meeting summarization microservice on Azure Container Apps, plus the transcript acquisition and ASR pipeline with Whisper fallback for meetings without published captions
- ▸Proposed and began building ASK NOHA, a hybrid RAG and knowledge-graph retrieval system over ~1,600 civic meetings, combining BM25, dense retrieval and Neo4j graph traversal with claim-level citation validation
Python
FastAPI
MongoDB
Neo4j
Weaviate
FAISS
Gemini
Whisper
Docker
3
Teaching Associate — Statistics
California State University – East Bay
Aug 2024 – Dec 2024
- ▸Taught probability, statistical inference, hypothesis testing and sampling to a cohort of 22 students through structured lessons and applied examples
- ▸Designed lectures, exercises and assessments connecting statistical concepts to real analytical questions
- ▸Explained quantitative methods to students with varied mathematical backgrounds, strengthening communication with non-specialist audiences
Statistics
Probability
Hypothesis Testing
Python
4
Founding Technical Advisor
Petpin AI · advisory engagement with a pre-funding founder
Feb 2024 – Aug 2024
- ▸Turned a founder's early pet-wearable concept into a defined product — shaping requirements, scoping technical feasibility on constrained hardware, and designing the end-to-end system architecture across device, sensing, connectivity and cloud
- ▸Built working prototypes on Raspberry Pi with the Pi AI Camera, running YOLO-based on-device object detection to automatically capture meaningful moments rather than recording continuously
- ▸Specified the sensing and hardware module architecture — camera, movement tracking, Wi-Fi connectivity — feeding an emotion- and activity-inference pipeline
- ▸Developed Python pipelines for real-time ingestion, preprocessing and TensorFlow Lite inference under strict device memory, battery and latency constraints
- ▸Designed the first physical device concept, including mounting approach and universal fit across pet sizes and collar types
- ▸Built React dashboards visualizing live telemetry, behavioral signals, device health and model outputs from connected hardware
Raspberry Pi
Pi AI Camera
YOLO
TensorFlow Lite
Python
React
5
Founder
R P Era · sole proprietorship, team of 4–5
Aug 2018 – Aug 2023
- ▸Founded and ran a five-line 3D-printing and laser-engraving business spanning workshops, hardware sales, repair services, imported machine distribution and custom manufacturing, managing a team of 4–5 and every operational function
- ▸Built the business's first customer-analytics workflow after inbound volume from IndiaMART reached hundreds of leads per day, moving from manual spreadsheets to Python-based analysis to identify which customer segments, product categories and channels actually converted
- ▸Established the India import channel for Ortur desktop laser engravers, handling supplier relationship, import logistics, pricing and downstream B2B and marketplace distribution
- ▸Delivered 100+ customer projects across engineering, dental and corporate custom-gift work, translating vague requirements into specifications, material decisions, production plans, estimates and finished parts
- ▸Built internal tooling for quotations, job tickets, inventory, invoicing and production tracking, and automated repetitive CAD-file preparation for engraving and printing workflows
Python
Flask
SQL
CAD Automation
3D Printing
6
M.S. Statistics — Data Science Concentration
California State University – East Bay
2023 – 2025
- ▸GPA: 3.9 — recipient of the Heebok Park Scholarship
- ▸Data Science concentration: statistical modeling, machine learning and applied analytics
- ▸Teaching Associate for undergraduate statistics while completing graduate studies
Statistical Modeling
Machine Learning
Python
R
7
B.E. Mechanical Engineering
University of Mumbai
2014 – 2018
- ▸Bachelor of Engineering in Mechanical Engineering
Mechanical Engineering
CAD/CAM
Founding AI Engineer
OpGov.ai · an initiative of Lead for Earth (non-profit)
Aug 2025 – Present
- ▸Took over as technical owner of the entire platform after the lead developer departed, growing ASK NOHA into an agentic system orchestrated with LangGraph and exposed as governed tools through an MCP server, with human-in-the-loop approval on sensitive actions and a policy layer enforcing autonomy boundaries
- ▸Developed a versioned evaluation harness with bootstrap confidence intervals on retrieval deltas, ablation studies, and an LLM-as-judge calibrated against human labels via Cohen's kappa — used as a release gate, not a report
- ▸Shipped four more production systems end to end: a publishing platform with a role-based permissions model I designed, a multi-channel campaign communications platform used in a real election, and a geospatial field operations dashboard used in a real campaign
- ▸Engineered a canonical MongoDB platform unifying 1.7M+ public-record profiles and ~4M historical participation records, reconciling a 734,457-record snapshot into 20,060 new profiles while isolating a single ambiguous identity for review
- ▸Administer the platform's infrastructure end to end — the public site on Vercel, the summarization service on Azure Container Apps, and everything else, including ASK NOHA's agent runtime and MCP server, on a Linux server I run myself
- ▸Own product design, system architecture, implementation, deployment and operations across the platform, coordinating a team of up to four including a remote developer
Python
LangGraph
MCP
Next.js
MongoDB
Neo4j
Weaviate
FAISS
Gemini
Azure
Linux
1
2
AI Engineer
OpGov.ai
Jan 2025 – Aug 2025
- ▸Joined as the platform's AI engineer and built its first AI service: the meeting summarization microservice on Azure Container Apps, plus the transcript acquisition and ASR pipeline with Whisper fallback for meetings without published captions
- ▸Proposed and began building ASK NOHA, a hybrid RAG and knowledge-graph retrieval system over ~1,600 civic meetings, combining BM25, dense retrieval and Neo4j graph traversal with claim-level citation validation
Python
FastAPI
MongoDB
Neo4j
Weaviate
FAISS
Gemini
Whisper
Docker
Teaching Associate — Statistics
California State University – East Bay
Aug 2024 – Dec 2024
- ▸Taught probability, statistical inference, hypothesis testing and sampling to a cohort of 22 students through structured lessons and applied examples
- ▸Designed lectures, exercises and assessments connecting statistical concepts to real analytical questions
- ▸Explained quantitative methods to students with varied mathematical backgrounds, strengthening communication with non-specialist audiences
Statistics
Probability
Hypothesis Testing
Python
3
4
Founding Technical Advisor
Petpin AI · advisory engagement with a pre-funding founder
Feb 2024 – Aug 2024
- ▸Turned a founder's early pet-wearable concept into a defined product — shaping requirements, scoping technical feasibility on constrained hardware, and designing the end-to-end system architecture across device, sensing, connectivity and cloud
- ▸Built working prototypes on Raspberry Pi with the Pi AI Camera, running YOLO-based on-device object detection to automatically capture meaningful moments rather than recording continuously
- ▸Specified the sensing and hardware module architecture — camera, movement tracking, Wi-Fi connectivity — feeding an emotion- and activity-inference pipeline
- ▸Developed Python pipelines for real-time ingestion, preprocessing and TensorFlow Lite inference under strict device memory, battery and latency constraints
- ▸Designed the first physical device concept, including mounting approach and universal fit across pet sizes and collar types
- ▸Built React dashboards visualizing live telemetry, behavioral signals, device health and model outputs from connected hardware
Raspberry Pi
Pi AI Camera
YOLO
TensorFlow Lite
Python
React
Founder
R P Era · sole proprietorship, team of 4–5
Aug 2018 – Aug 2023
- ▸Founded and ran a five-line 3D-printing and laser-engraving business spanning workshops, hardware sales, repair services, imported machine distribution and custom manufacturing, managing a team of 4–5 and every operational function
- ▸Built the business's first customer-analytics workflow after inbound volume from IndiaMART reached hundreds of leads per day, moving from manual spreadsheets to Python-based analysis to identify which customer segments, product categories and channels actually converted
- ▸Established the India import channel for Ortur desktop laser engravers, handling supplier relationship, import logistics, pricing and downstream B2B and marketplace distribution
- ▸Delivered 100+ customer projects across engineering, dental and corporate custom-gift work, translating vague requirements into specifications, material decisions, production plans, estimates and finished parts
- ▸Built internal tooling for quotations, job tickets, inventory, invoicing and production tracking, and automated repetitive CAD-file preparation for engraving and printing workflows
Python
Flask
SQL
CAD Automation
3D Printing
5
6
M.S. Statistics — Data Science Concentration
California State University – East Bay
2023 – 2025
- ▸GPA: 3.9 — recipient of the Heebok Park Scholarship
- ▸Data Science concentration: statistical modeling, machine learning and applied analytics
- ▸Teaching Associate for undergraduate statistics while completing graduate studies
Statistical Modeling
Machine Learning
Python
R
B.E. Mechanical Engineering
University of Mumbai
2014 – 2018
- ▸Bachelor of Engineering in Mechanical Engineering
Mechanical Engineering
CAD/CAM
7