---
title: "Selected work | Dhruv Doshi"
description: "Selected AI systems, platform engineering, distributed systems, observability, and product work by Dhruv Doshi."
canonical: https://doshidhruv.com/projects/
author: Dhruv Doshi
---

[Skip to content](#main-content)

[Dhruv Doshi](/)

[Work](/projects)[Experience](/resume)[Notes](/notes)[Guides](/guides)[Research](/research)[About](/about)[Contact](/contact)[Search /](/search)dark mode

# Work

## Architecture Solution Blueprint platform

Designed and led the zero-to-one delivery of an enterprise platform that turns 1,000+ architecture patterns into real-time service recommendations, solution blueprints, and automated architecture workflows.

- Designed the rules-based recommendation and auto-certification engine.
- Built the core recommendation logic and most of the production implementation.
- Mapped security, integration, infrastructure, and deployment patterns to enterprise services.
- Built for air-gapped, security-constrained enterprise environments.
- Led five engineers through ADRs, reviews, delivery, and stakeholder alignment.

**Technologies:** TypeScript, React, Node.js, Rules engine, Enterprise architecture

- [FINOS CALM and architecture as code](/notes/finos-calm-and-architecture-as-code)
- [Production RAG requires retrieval evidence and control](/notes/production-rag-requires-retrieval-evidence-and-control)
- [Pattern matching algorithms for architecture recommendations](/notes/pattern-matching-algorithms-for-architecture-recommendations)
- [Platform as a Service (PaaS)](/notes/platform-as-a-service-paas)
- [What is cloud computing?](/notes/what-is-cloud-computing)
- [The Fundamentals of Machine Learning](/notes/the-fundamentals-of-machine-learning)

## Vendor-neutral observability platform

Architected a vendor-neutral OpenTelemetry and telemetry-routing platform designed for approximately 150 TB/day across 10,000+ systems, enabling governed collection and routing without coupling teams to a single observability vendor.

- Defined OpenTelemetry schemas for institutional logs and metrics.
- Combined Vector, Fluent Bit, Kafka, and Logstash into reusable collection paths.
- Created SDK guidance and onboarding contracts adopted by 20+ engineering teams.
- Designed Microsoft 365 audit-log repatriation with geolocation-aware data handling.

**Technologies:** OpenTelemetry, Kafka, Vector, Fluent Bit, Logstash

- [OpenTelemetry pipeline architecture for vendor-neutral observability](/notes/opentelemetry-pipeline-architecture-for-vendor-neutral-observability)
- [Microsoft 365 audit data in an OpenTelemetry pipeline](/notes/microsoft-365-audit-data-in-an-opentelemetry-pipeline)
- [AWS support for OpenTelemetry with ADOT and CloudWatch](/notes/aws-support-for-opentelemetry-with-adot-and-cloudwatch)
- [Splunk support for OpenTelemetry collection and OTLP](/notes/splunk-support-for-opentelemetry-collection-and-otlp)
- [Datadog support for OpenTelemetry and OTLP ingestion](/notes/datadog-support-for-opentelemetry-and-otlp-ingestion)
- [Dynatrace support for OpenTelemetry and native OTLP](/notes/dynatrace-support-for-opentelemetry-and-native-otlp)
- [Elastic support for OpenTelemetry and managed OTLP ingestion](/notes/elastic-support-for-opentelemetry-and-managed-otlp-ingestion)
- [Distributed cloud](/notes/distributed-cloud)
- [Multi-cloud architecture](/notes/multi-cloud-architecture)
- [Serverless computing and Function as a Service](/notes/serverless-computing-and-function-as-a-service)

## ASB Assist

Building an LLM and RAG architecture assistant that turns natural-language requirements into architecture recommendations and diagrams using retrieval, tool integration, evaluation, grounding controls, guardrails, and auditable workflows.

- Designing permission-aware retrieval and citation validation for architecture knowledge.
- Building offline evaluation for recommendation quality, grounding, and failure behaviour.
- Applying guardrails and approval boundaries before generated output can influence architecture decisions.
- Exploring MCP and agent-gateway patterns for controlled tool access.

**Technologies:** Python, LLMs, RAG, Evaluation, Tool integration

- [Production RAG requires retrieval evidence and control](/notes/production-rag-requires-retrieval-evidence-and-control)
- [Design safe tool use for AI agents](/notes/design-safe-tool-use-for-ai-agents)
- [Evaluate LLM systems as systems](/notes/evaluate-llm-systems-as-systems)
- [Define the boundaries of a production RAG system](/notes/define-the-boundaries-of-a-production-rag-system)

## Healthcare verification platform

Joined as the first engineer and built the platform across React and TypeScript, Node.js APIs, verification workflows, document-classification integration, and AWS EKS in a regulated healthcare environment.

- Designed HIPAA-conscious data handling and audit logging.
- Built a Hyperledger Fabric verification layer and ERC-20 contracts.
- Containerised the platform and automated deployment workflows.
- Integrated document-classification models for medical record processing.

**Technologies:** React, Node.js, AWS EKS, Hyperledger Fabric, Solidity

- [What is Blockchain?](/notes/what-is-blockchain)
- [Hybrid cloud](/notes/hybrid-cloud)
- [Infrastructure as a Service (IaaS)](/notes/infrastructure-as-a-service-iaas)

## Research and independent projects

### DCS-BBN

A peer-to-peer archival storage network using Solidity smart contracts, Hyperledger Fabric, and privacy-preserving verification. Published by Springer.

### Automatic trading system

A low-latency trading system that combines live brokerage APIs, market-data streams, quantitative signals, position sizing, and automated risk controls.

**Dhruv Doshi** · Toronto, Canada · [work@doshidhruv.com](mailto:work@doshidhruv.com)

[Resume](/resume)[Notes](/notes)[Guides](/guides)[Topics](/topics)[Search](/search)[Research](/research)[LinkedIn](https://www.linkedin.com/in/dhruvdoshi25071999)[GitHub](https://github.com/DhruvDoshi)

[Sitemap](/sitemap.xml)[RSS](/feed.xml)[LLMs](/llms.txt)

© 2026 Dhruv Doshi