Zachary Proser

HOVER PREVIEW · CLICK TRAVEL · T TOUR · ← → BROWSE · ENTER READ

§ 00 · Mind on Fire LLC — Software & AI consultancy

The mind is a fireto be ignited.

I’m Zachary — a builder, writer, and teacher.

The fire above draws from my 182 essays. Explore the full constellation on a larger screen, or browse every essay below.

✱The Modern Coding letter5,000+ engineers
Dispatches from the edge of applied AI.

What lands in your inbox: what I’m building, what I’m learning, the tools I’d actually pay for. Names named, vendors graded, evals included.

Unsubscribe in one click · No spam, ever
Read all my writing →·Workshops & AI enablement →
ZRead essay →
Who's writing
Portrait of Zachary Proser
Plate I · MMXXVI
I do this work because I love it — and everything I learn, I give away.

Hi, I'm Zachary.

I work on the Applied AI team at WorkOS, shipping internal and customer-facing systems, automations, and apps that cut toil — and bringing the rest of the org along with workshops on AI fundamentals, training, office hours, and support.

I'm also an independent AI researcher working on local inference, inference engineering, and cost optimization. Through my consultancy, Mind on Fire, I do custom workshops, software and systems development, and bespoke training — on my own platform and as a Hone coach. For the last two years I've taught engineers everything from inference to agentic development, loops, scheduling, and skills at scale at the AI Engineer conferences in London, New York, and San Francisco.

More about me →·Say hello →
Currently
Applied AI · WorkOS
Consultancy
Mind on Fire
Researching
Local inference, inference engineering, cost optimization
17
Blueprint drawings
60+
Controlled benchmark runs
6,000+
Newsletter subscribers
12.4M
Search impressions / yr
§ 01

Research

All research →

Controlled experiments in local inference economics, eval methodology, and serving architecture — every claim backed by runs you can inspect.

The Price Floor
Local inference economicsFlagship

The Price Floor

Break-even analysis: M5 Max memory-bandwidth ceilings, the engine tax (llama.cpp vs ds4 = 5×), and the ~$34K Blackwell build-or-buy.

The Eval Harness
Eval methodology60 runs

The Eval Harness

Coding agent benchmark — local vs cloud vs frontier, hidden-test design, Wilson interval statistics.

EconomicsThe Cost CurveDeepSeek V4 Flash: MoE routing, quantization ladder, 96.6% cache-hit dominance.ArchitectureThe Inference EnginePrefill/decode, KV cache & PagedAttention, SARATHI, DistServe.MethodologyThe BenchmarkMMLU, SWE-bench, Chatbot Arena — and Goodhart’s law in eval.Field guidePractical Local InferenceM5 Max + DeepSeek V4 Flash at 30–40 tok/s. $17/mo via Gateway.
§ 02

The Blueprint Series

All 17 →

17 technical drawings of the AI stack — SVG schematics, equations, citations, interactive demos.

The TransformerThe Transformer№ 01The Attention HeadThe Attention Head№ 02The TokenizerThe Tokenizer№ 03The Embedding SpaceThe Embedding Space№ 04The RAG PipelineThe RAG Pipeline№ 05+12View all →
§ 03 · Workshops · From $15kQ4 open

Taught from the bench, not the slide deck.

The workshop material comes straight from the research above — real cost curves, evals your team can rerun. We embed inside your pods and leave you with working workflows and named champions. Built with Anthropic; 800 registrations for the last public run.

Book a workshop →Engagement shapes →
~/workshops — calendarLive

Booking the next collaboration weeks.

Jun 17Healthcare · Org-wideBooked
Jul 22Fintech · Two-dayBooked
Aug — openYour org here→
Sep — openYour org here→
Lead time 6–10 wkHold a date →
§ 04

Interactive demos

All 9 →
Local vs Cloud BenchmarkLocal vs Cloud Benchmark60-run evidence explorer.Tokenization LabTokenization LabBPE pricing calculator.RAG VisualizedRAG VisualizedPipeline step-through.Embeddings ExplorerEmbeddings Explorer3D vector space.

§ 05Publications

Full list →
arXiv · 2510.25819 · Oct 2025
Identity Management for Agentic AI
With Tobin South, Alex Pentland (MIT), and others.
ChapterRAG Evaluation — official contribution to Pinecone’s Vector Databases in Production book series.
50+ articlesTechnical writing at WorkOS, Pinecone, Cloudflare, Gruntwork — agents, RAG, auth, vector DBs, IaC.
180+ essayszackproser.com/blog, plus The New Stack, Prefect.io, and the Pinecone Learning Center.
§ 06

On stage

All talks →
Untethered ProductivityAI Engineering London · June 2026Untethered Productivity
Modern AI and SecurityDevSecCon 2025 · KeynoteModern AI and SecurityClaude Cowork GTM WorkshopWorkOS × Anthropic · WorkshopClaude Cowork GTM WorkshopSkills at ScaleAIE London · 80-min workshopSkills at ScaleFrom Jupyter Notebooks to Productiona16z SF · TalkFrom Jupyter Notebooks to Production
§ 07

Career

14+ Years Engineering Experience
WorkOSApplied AI2025–nowAI workflows, internal tooling, developer education. Created the Claude Cowork workshop with Anthropic.
WorkOSDeveloper Education2024–25Built and led the developer-education surface at WorkOS — documentation, workshops, and video.
PineconeStaff DevRel2023–24RAG pipelines, AWS reference architectures, dev education during the vector DB explosion.
GruntworkTech Lead2020–23Built git-xargs (1K★), contributed to cloud-nuke (3.1K★). Multi-account AWS, production IaC.
CloudflareSenior SWE2017–20Joined at ~100 engineers. Built cf-terraforming (1.3K★). APIs, edge infra, dev tools at scale.
12.4M
Search impressions / yr
52.5K
Search clicks · avg pos 7.9
137K
Visitors / yr · +21% YoY
179K
Page views / yr · +18%
6,000+
Newsletter subscribers

Regularly advising hedge funds and PE firms on AI infrastructure via AlphaSights, Guidepoint, Tegus, GLG, Coleman.

🔥 Mind on Fire — independent AI consultancy
Local inference economics, eval methodology, and systems from silicon to prompt.
mindonfire.net →