Derek Shaw

Software Developer

19years in tech

8years leading teams

6/10Data Layer & Database

8years with C#

Overview

Senior Rust systems engineer specializing in developer tooling and code-indexing pipelines. The strongest proven skill is building robust, concurrent Rust CLI services and code-index loaders as evidenced by engraph/crates/engraph-codegraph/src/scip_loader.rs and engraph/crates/engraph-cli/src/main.rs with accompanying unit and integration tests. There is no public evidence of embedded firmware, device drivers, PCB/FPGA work, or on-device HIL testing in the code analyzed.

Phone

Timeline

Senior Software Engineer, Growth · Senior

City Storage Systems · Full-Time

Mar 2026 to Present 8 Months Sammamish In office

Built a lead generation platform for the sales team using an on-demand enrichment pipeline exposed via gRPC and MCP endpoints. Addressed data gaps in physical infrastructure by implementing a PDF parser for extracting information from architectural drawings. Led observability initiatives with OpenTelemetry by defining instrumentation patterns, shipping tracing across services, and onboarding engineers to the approach. Developed an agentic AI QA/release system using Claude, working with Rust, Python, and Java.

gRPC

Model Context Protocol

OpenTelemetry

Rust

Python

Java

Claude

Principal Software Engineer / Architect · Lead

Lynk Global · Full-Time

Aug 2025 to Mar 2026 7 Months Chantilly Partially remote

Designed and implemented OpenTelemetry integrations in Rust for the MOS microservices platform running on AWS. Built a self-hosted observability pipeline using Grafana Tempo, Mimir, and Loki with the AWS Distro for OpenTelemetry collector to cover traces, metrics, logs, and derived span metrics. Created a multi-language telemetry bootstrap library to standardize SDK initialization and exporters for new services, and helped troubleshoot metric temporality issues across components.

Rust

OpenTelemetry

AWS

Amazon ECS

AWS Fargate

Tempo

Mimir

Loki

TypeScript

C++

Python

Principal Engineer · Lead

Microsoft · Full-Time

Jan 2018 to Nov 2024 6 Years 10 Months Bellevue In office

Owned two core Office File IO systems used across internal traffic routing and endpoint federation for government and other external integrations. Ensured cross-platform compatibility across Windows, macOS, iOS, and Android clients. Produced API documentation and operational monitoring dashboards, tracked status to senior leadership, and refactored implementations to improve reliability and performance. Worked with customers to onboard to new API surfaces and adopt recommended usage practices.

Principal Engineering Manager · Lead

Microsoft · Full-Time

Nov 2015 to Jan 2018 2 Years 2 Months Bellevue In office

Managed a team of software developers building Microsoft Excel experiences for Apple products including iOS, macOS, and related components. Acted as scrum master, ran business reviews, triaged issues, and coached the team for performance and delivery. Directed an accessibility initiative driven by new government requirements, helping ship on time while meeting elevated accessibility standards and mitigating regulatory risk.

Scrum

Principal Development Manager · Lead

Microsoft · Full-Time

Jan 2013 to Oct 2015 2 Years 9 Months Bellevue In office

Created product experiences for Bing and led a team of UX developers responsible for ideation tooling and search-related workflows. Partnered with researchers to run user studies, produce prototypes and live mock-ups, and iterate solutions based on findings. Conceived and delivered a prototype integration concept for the Windows start bar in collaboration with the Windows team.

Principal SDE Lead · Lead

Microsoft · Full-Time

Jan 2010 to Jan 2013 3 Years Bellevue In office

Led and coached a group of developers for the Bing Segments UX team. Coordinated cross-functional collaboration to ship user experiences across content and entertainment segments such as people, celebrities, movies, and more. Managed planning, daily triage, and contributed to a mid-tier architecture that enabled onboarding UX experiences with minimal UX team involvement.

Senior Director of Development / Architecture · Head+

FunMobility · Full-Time

Feb 2007 to Aug 2008 1 Year 6 Months Pleasanton In office

Reported to the CTO and led multiple projects for a mobile marketing solutions company. Developed a new architecture using Java, Hibernate, and SQL Server for mobile/WAP and web deployments, improving performance and resource usage. Drove delivery and engineering direction across the development efforts during the role.

Java

Hibernate

MS SQL

Software Engineer · Middle

Claris · Full-Time

In office

Software engineering role at Claris (an Apple company). The CV does not provide specific responsibilities, technologies, or outcomes for this position.

Software Engineer · Middle

Adobe · Full-Time

In office

Software engineering role at Adobe. The CV does not provide specific responsibilities, technologies, or outcomes for this position.

Software Engineer III · Middle

Electronic Arts (EA) · Full-Time

In office

Software engineering role at Electronic Arts. The CV does not provide specific responsibilities, technologies, or outcomes for this position.

Office Experimentation Tech Lead · Lead

Microsoft · Full-Time

Bellevue In office

Provided technical leadership and design for an SDK consumed across Office applications, focusing on native C++ plus Java, Objective-C, and Swift clients. Authored design documentation and guided engineers through implementation of a new API that improved performance by over 30%. Led refactoring work to improve boot and in-use performance while simplifying maintenance and reducing resource usage.

C++

Java

Objective-C

Swift

Office Diagnostics Tech Lead · Lead

Microsoft · Full-Time

Bellevue In office

Led technical and delivery planning for Office diagnostics, including team workload management and roadmap execution. Migrated a native C++ Windows diagnostics system to run on macOS and iOS by creating a shared codebase. Enabled cross-platform parity so fixes could flow both directions between Windows and Apple platforms.

C++

Office Signals Tech Lead · Lead

Microsoft · Full-Time

Bellevue In office

Improved engineering automation for Office by building infrastructure to collect traces during automation workflows using Azure and C#/.NET. Increased code coverage and improved test selection for Office, including PR and CI/CD-driven test execution. Used telemetry-driven practices within the Office automation architecture to accelerate integration for thousands of developers.

Azure

Python

C++

CI/CD

Technical skills

Languages

Java

C++

Python

TypeScript

Rust

SQL

JavaScript

Node JS

Swift

Objective-C

Python

FastAPI

Pydantic

SQLAlchemy

Rust

Clap

Actix Web

SQLx

Tracing

Tokio

DevOps

Azure

CI/CD

Amazon ECS

AWS Fargate

OpenTelemetry

Loki

AWS

Other

Hibernate

MS SQL

OpenAI Codex

Tempo

PostgreSQL

React.js

STL

MySQL

Ray

Claude

Rest API

Mimir

Bootstrap

Vue.js

GitHub

Grafana

Couchbase

SQLite

Uvicorn

Mypy

Git

Linux

Windows

Model Context Protocol

Apache Tomcat

gRPC

Apache HTTP Server

Ubuntu

Unix

AI Agents

Debian

Microsoft Excel

Middle Backend Developer Confidence: High API Engineer

Senior backend engineer specializing in API-driven systems and developer tooling with a strength in Rust-based server and indexing code. The strongest proven skill is building robust backend data and retrieval layers, evidenced by the GraphQL API and sqlx-backed models and tests in backend/src/graphql.rs and backend/src/db_models.rs. Public code shows good test coverage and modular design but limited evidence of large-scale distributed system patterns, formal token/auth lifecycle implementations, or cloud native ops (k8s/infra pipelines).

API Design

5/10

How well APIs are designed

API design shows deliberate GraphQL schema and test coverage but lacks explicit versioning, idempotency key patterns, and advanced pagination contracts.

Evidence

derekdshaw/transaction_server/backend/src/graphql.rs: QueryRoot and MutationRoot GraphQL schema and conversions derekdshaw/transaction_server/backend/tests/gql_transaction_tests.rs: end-to-end GraphQL query/mutation tests using create_schema

Data Layer & Database

6/10

Working with databases

Data layer shows careful SQL usage, transactional boundaries, and migration/test artifacts; evidence of explicit transactional swap/rollback patterns and prepared queries.

Evidence

derekdshaw/transaction_server/backend/src/db_models.rs: sqlx query_as! usage and transactional CTEs for create/update derekdshaw/transaction_server/backend/tests/db_transaction_tests.rs: integration tests using test DB to verify transactional behavior

Scalability & Performance

5/10

Handling load and speed

Scalability and performance considerations are present (async runtime, sqlx, tests with containers) but there are few documented, production-grade scaling artifacts like distributed queues, measured load-tests, or caching/invalidation strategies.

Evidence

derekdshaw/transaction_server/backend/Cargo.toml: async stack (tokio) and sqlx dependencies derekdshaw/transaction_server/backend/tests/common/test_utils.rs: testcontainers-driven integration test utilities

System Architecture

6/10

Overall system structure

Clear modular decomposition (separated crates/services, CLI/ingest/retrieve layers in other projects) and conscious architecture decisions for indexing and service boundaries; service decomposition is pragmatic rather than microservices-for-show.

Evidence

derekdshaw/transaction_server/backend/README.md: project-level architecture and separation (backend, frontend, classifier) derekdshaw/transaction_server/backend/src/graphql.rs: separation of GraphQL context and repository traits

Security & Auth

4/10

Protecting data and access

Reasonable input-sanitization and parameterized queries limit injection risk; however there is limited evidence of comprehensive auth/token lifecycle, secret management or hardened SSTI/SSRF mitigations in the public code.

Evidence

derekdshaw/transaction_server/backend/src/db_models.rs: use of sqlx query_as! and typed parameters derekdshaw/transaction_server/backend/src/config.rs: dotenv-based environment config (loader for env variables)

Reliability & Observability

5/10

Stability and monitoring

Observability and reliability are present via structured logging and tests; handler timeouts and integration test fixtures exist but advanced resilience patterns (circuit breakers, backoff strategies with reasoning) are limited in the visible code.

Evidence

derekdshaw/transaction_server/backend/Cargo.toml: includes env_logger for logging derekdshaw/transaction_server/backend/tests/db_transaction_tests.rs: integration tests validating DB behavior and ordering

Expertise

Backend AI & LLM· Middle

Rust· Middle

Python· Middle

Microservices & API Architecture· Middle

Industries

Artificial Intelligence· Middle

Financial Services· Middle

Software· Middle

Technologies

Python· since 2025 · Middle

Rust· since 2025 · Middle

Rest API

gRPC· since 2026

SQLAlchemy

Hibernate· since 2007

FastAPI

Tokio

Actix Web

Tracing

SQLx

Pydantic

Mypy

Uvicorn

Clap

Persistent· mentioned only

Telemetry· mentioned only

Recommendations

Lead development of backend APIs and GraphQL services that require strong type safety and SQL-backed transactions (design and implement complex queries and migrations).
Develop developer tooling and code-indexing/retrieval features (local-first retrieval, SCIP-based indexing, and compact output filters) for LLM integrations and IDE tooling.
Build Python-based LLM-backed agent services (FastAPI endpoints that call local or external LLMs) integrating with existing SQL stores and recommendation pipelines.
Implement database-focused features such as robust migration chains, integration tests with testcontainers, and production-grade SQL performance tuning and indexing.

Repositories

The developer's experience in this domain has been verified based on AI analysis of the following repositories:

engraph mnemosyne transaction_server

Middle Embedded Engineer Confidence: High Generalist

Embedded & Firmware

Low-level device code

Not evidenced in public code

Hardware-Software Interface

Connecting code to hardware

Not evidenced in public code

Resource Constraints

5/10

Working with limited resources

Resource-aware design is evident in careful concurrency throttling, avoidance of unnecessary heap allocations, and database pragmas/migration handling. The code shows deliberate choices to limit parallelism, bound handler time, and reduce allocations in hot paths rather than naive defaults.

Evidence

mnemosyne/memory-mcp-server/src/main.rs: Param enum used to avoid heap allocations and run_db_with_timeout implements bounded blocking work

git_dag_analyzer/src/git_processing.rs: build_deps_tokio uses Semaphore, JoinSet, and Atomic counters to limit and track concurrent git subprocesses

engraph/crates/engraph-core/src/db.rs: open_db/open_pool and setup_pragmas functions managing SQLite pragmas and WAL/migration behavior

Real-time & Timing

5/10

Precise timing control

Clear handling of concurrency and timing: signal handling, child process I/O draining to avoid pipe deadlocks, handler timeouts, and explicit progress/timing reporting demonstrate attention to latency and liveliness rather than naive blocking.

Evidence

engraph/crates/engraph-cli/src/main.rs: run_wrapped_command reads stdout and stderr concurrently and installs best-effort signal handlers to avoid deadlocks

git_dag_analyzer/src/git_processing.rs: build_deps_tokio implements a progress reporting task and limits concurrency with a semaphore

mnemosyne/memory-mcp-server/src/main.rs: HANDLER_TIMEOUT and run_db_with_timeout enforce time bounds on tool handlers

HDL & Circuit Logic

Designing digital circuits

Not evidenced in public code

Reliability & On-device Testing

5/10

Testing on real hardware

Strong software reliability practices: transactional RAII, rollback-on-error, structured logging, defensive DB handling, and wide unit/test coverage for critical parsing and merge logic. There is no evidence of on-device hardware-in-the-loop testing or firmware watchdogs, but the server-and-CLI reliability story is well covered.

Evidence

engraph/crates/engraph-codegraph/src/scip_loader.rs: TxGuard RAII for rollback-on-drop, prune_stale_symbols and many unit tests around load/merge behavior

engraph/crates/engraph-cli/tests/*: multiple integration/unit tests for hooks and command rewriting demonstrating test coverage for critical features

mnemosyne/memory-mcp-server/src/main.rs: poisoned-mutex handling and structured error-to-MCP conversion (LogErr trait) plus explicit handler timeouts

Expertise

Low-level Development & Tooling· Middle

Recommendations

Develop high-performance developer tools and language-server/indexer integrations (CLI-driven Rust services that spawn and manage external indexers).
Build code-indexing, retrieval, and telemetry backends that require careful concurrency control and SQLite-based local storage.
Implement reliable stdio/rpc daemons and hook handlers that must be robust to malformed input and bounded in runtime (MCP-style servers).
Lead performance-sensitive Rust work such as parallel repository analysis and subprocess orchestration where resource limits and progress semantics matter.

Repositories

The developer's experience in this domain has been verified based on AI analysis of the following repositories:

engraph mnemosyne git_dag_analyzer

Junior Game Developer Confidence: Medium Generalist

Middle-level software developer specializing in low-level data structures and performance-focused Rust code. Their strongest proven skill is implementing and benchmarking performance-sensitive data structures, demonstrated by the Rust BST implementation and Criterion benches (bst_testing/rust/src/bst.rs and bst_testing/rust/benches/bst_benchmark.rs). There is no public evidence of game-engine work, rendering/shader development, netcode, or editor tooling in the human-authored artifacts.

Gameplay Systems & Mechanics

How game logic works

Not evidenced in public code

Graphics & Rendering

Drawing game visuals

Not evidenced in public code

Physics & Math

2/10

Game physics and math

Algorithmic/data-structure work is present (correctness and edge-case handling in BST operations) but no physics/simulation integrators or spatial partitioning specific to game physics.

Evidence

bst_testing/rust/src/bst.rs: delete_internal / find_min_idx implementations and associated correctness tests

bst_testing/cpp/bst.cpp: DeleteNode_Internal / minValueNode algorithm implementations

Engine Proficiency

Skill with the game engine

Not evidenced in public code

Performance & Frame Budget

4/10

Keeping the game smooth

Clear attention to performance tuning and measurement: vector-backed node storage to reduce allocations, dedicated release-performance profile settings, and Criterion benchmarks for large-data workloads.

Evidence

bst_testing/rust/src/bst.rs: nodes stored in Vec to reduce allocations (alloc_node)

bst_testing/rust/Cargo.toml: release-performance profile with LTO and profiling-friendly settings

bst_testing/rust/benches/bst_benchmark.rs: Criterion benchmarks for build/delete on large datasets

Content Pipeline & Tooling

3/10

Tools for game content

Tooling and pipeline-level attention for a small systems project: bench harness, Cargo profile tuning and test coverage; no broad content pipelines or CI for large asset/tool chains.

Evidence

bst_testing/rust/Cargo.toml: custom release-performance profile and bench config

bst_testing/rust/benches/bst_benchmark.rs: benchmark group configuration and long measurement runs

bst_testing/README.md: describes test/benchmark workflow across languages

Technologies

C++· mentioned only

Recommendations

Develop performance-sensitive library code and systems utilities in Rust (low-level data structures, allocators, or core algorithms).
Implement and profile algorithmic components where predictable low-latency and memory behavior matters (bench-driven optimization and microbenchmark ownership).
Port or harden core algorithms for engine subsystems (pathfinding, scene indexing, resource managers) that need vetted correctness and benchmarks.

Repositories

The developer's experience in this domain has been verified based on AI analysis of the following repositories:

engraph mnemosyne bst_testing

Job preferences

Work permits: United States
Citizenship: United States
Employment type: Full-Time

Open to AI agents: they read this profile and start a dialog through Alionalion.io/@derek-shaw/mcp