Vibe Engines
YouTube
HANDBOOKS · READ THEM IN ORDER

The complete handbook library.

208 full-length technical handbooks — AI engineering, system design, forward-deployed work and career — arranged as 27 reading ladders rather than a wall of cards. Each rung assumes the one below it, and every "X vs Y" comparison sits directly after both things it compares, because that is the only point at which it is answerable.

208Handbooks
27Ladders
55hOf reading
100%Free · no sign-up
Start hereThe Prompting Handbook~45 min — the only handbook in the library that assumes nothing before it, and rung 1 of the Prompting and context ladder. If you want the engineering floor instead, start at ML Fundamentals or System Design Fundamentals, each rung 1 of its own ladder below.Read it →
208 handbooks

The machine underneath

What every later handbook quietly assumes: how a program becomes instructions, where data actually sits, and how two machines talk.

How computers actually work

Seven handbooks that take you from "what does a compiler do" to reading a stack trace and knowing which layer lied. Every systems handbook below assumes this floor.

0 / 7 read · 1h 46m
  1. How Programming Languages Actually Work Start here Four questions of timing: when your code gets translated, when type errors get caught, what actually gets copied on a call, and what a…~20 min
  2. The Compilers Basics HandbookHow a flat string like "2 + 3 * 4" becomes the meaning 14 (not 20).~16 min
  3. The Operating Systems Fundamentals HandbookHow your machine runs hundreds of programs on a few cores and a fixed slab of RAM — by sharing what isn't enough.~16 min
  4. Process vs ThreadProcess vs thread, decided by whether they share memory: a process is an independent program with its own isolated memory; a thread run…~9 min
  5. The Linux Internals HandbookLinux runs the world on one idea: everything is a file.~16 min
  6. The Browser Internals HandbookHow a browser turns a tree of tags and a pile of CSS into a scrolling, animating picture at 60fps.~16 min
  7. The Git Internals HandbookLearn the model and the commands become obvious.~13 min

The languages you ship in

Python first because everything in the AI arc is written in it, then the typed and compiled languages you end up next to. The interview guides double as a syllabus.

0 / 7 read · 2h 59m
  1. 50 Python Interview Questions Start here Fundamentals to advanced: data structures, OOP, iterators & generators, the GIL, asyncio, memory, testing, and the standard library — a…~50 min
  2. The TypeScript HandbookWhy TypeScript surprises people from Java/C#: its type system is structural, not nominal — it matches by an object's shape, not a decla…~15 min
  3. The React HandbookReact lets you write UI declaratively — return a description of what it should look like — and makes it fast with reconciliation: it bu…~16 min
  4. The React Interview HandbookWhat senior React interviews actually probe — components and JSX, props vs state, the render cycle with the virtual DOM and reconciliat…~16 min
  5. 50 Angular Interview QuestionsA visual handbook covering components, change detection, RxJS, signals, routing, forms, performance, and testing — what interviewers ac…~50 min
  6. The Go HandbookGo makes concurrency approachable with one motto: don't communicate by sharing memory; share memory by communicating.~16 min
  7. The Rust HandbookRust refuses the old memory trade-off (safe-but-slow GC vs fast-but-dangerous manual free) with one idea: ownership.~16 min

Data, and where it lives

SQL, then what Postgres is doing underneath it, then the three choices everybody argues about — relational or not, row or column, cache or not.

0 / 9 read · 1h 40m
  1. The SQL Handbook Start here What the database does underneath your SELECT — the relational model and keys, joins (inner/left/right/full), indexes and B-trees, the…~16 min
  2. The PostgreSQL Internals HandbookKey handbookUPDATE never updates, and a janitor keeps it working.~13 min
  3. SQL vs NoSQLFramed as a war, it’s really a menu.~9 min
  4. PostgreSQL vs MySQLBoth are excellent free relational databases you can build a company on, so the choice is about character, not capability: Postgres pri…~8 min
  5. OLTP vs OLAPOne storage-layout decision explains the whole comparison: row-oriented OLTP makes fetching one record cheap, column-oriented OLAP make…~8 min
  6. Optimistic vs Pessimistic LockingOptimistic vs pessimistic locking, decided by how likely two writers collide: pessimistic locks a row before editing so others wait; op…~9 min
  7. The Redis HandbookRedis as an in-memory data-structure server, not just a cache — the core structures (strings, hashes, lists, sets, sorted sets), why si…~14 min
  8. Redis vs MemcachedBoth put data in RAM by a key, so they look interchangeable — but Memcached does one thing (a lean multithreaded cache) while Redis is…~8 min
  9. The Caching Patterns HandbookWhy the fastest query is the one you never run.~15 min

Networks, APIs and auth

The wire, then the shapes people put on it, then the two things that go wrong in production: who is calling, and can they prove it.

0 / 9 read · 1h 51m
  1. The Networking Handbook Start here How a request actually travels — the layered model, IP addressing and routing, TCP vs UDP, the TCP handshake and head-of-line blocking,…~16 min
  2. TCP vs UDPTCP vs UDP, decided by whether every byte must arrive: TCP is connection-oriented and guarantees reliable, in-order delivery with flow…~9 min
  3. The API Design HandbookAn API is a promise you keep for years.~13 min
  4. gRPC vs RESTBoth let services talk, and you often want both.~8 min
  5. GraphQL vs RESTGraphQL vs REST, decided by who chooses the shape of the response: REST exposes many fixed resource URLs and the server decides what ea…~9 min
  6. REST vs Webhooks vs SSEThree ways client and server move data, split by who exposes an endpoint and whether the connection stays open: REST pulls, webhooks fl…~9 min
  7. The WebSockets & Real-Time HandbookWhy "live" is hard on a protocol that can't push.~15 min
  8. The OAuth & Auth Deep DiveHow "Sign in with Google" lets an app act for you without ever seeing your password.~16 min
  9. Cryptography for EngineersEncryption from XOR up.~16 min

Systems at scale

One box is never enough. Consistency, partitioning, message queues, and the operational work of keeping the result alive.

Distributed systems, from first principles

Start at the fundamentals handbook, then the four constraints every design interview is really testing, and close on the two consensus algorithms named in the answers.

0 / 7 read · 1h 50m
  1. The System Design Fundamentals Handbook Start here The load-bearing ideas behind every distributed system — the CAP theorem and consistency models, concurrency and locking, partitioning…~40 min
  2. CAP Theorem & Consistency ModelsWhat a distributed system can promise when the network splits — CP vs AP, why "CA" is a myth, PACELC, and the full spectrum from linear…~12 min
  3. Partitioning, Sharding & ReplicationHow one dataset becomes many — range vs hash partitioning, consistent hashing and virtual nodes, hot partitions, replication topologies…~13 min
  4. Concurrency, Locks & Isolation LevelsHow overlapping operations stay correct — race conditions, optimistic vs pessimistic locking, the four isolation levels and their anoma…~13 min
  5. Consensus, Transactions & CoordinationHow nodes that can crash still agree on one truth — majority quorums, Raft and Paxos, leader election, two-phase commit vs the saga pat…~14 min
  6. Raft vs PaxosProvably equivalent in power, radically different to implement: Raft bets on a strong leader to make consensus reasoning-friendly; Paxo…~9 min
  7. CP vs AP DatabasesCAP theorem’s forced choice, precisely stated: CP systems refuse some requests to guarantee consistency during a partition; AP systems…~9 min

Moving data between services

The log, the three brokers people pick between, and the two architecture arguments — batch or stream, one service or many — that the choice decides for you.

0 / 6 read · 57m
  1. The Kafka Handbook Start here Apache Kafka as a distributed append-only log, not a queue — topics, partitions and offsets, producers, consumers and consumer groups,…~14 min
  2. Kafka vs RabbitMQThey both "move messages", which is exactly why teams pick the wrong one.~9 min
  3. Kafka vs KinesisSame partitioned-log model underneath, different operational surface: Kafka is self-run and portable with a huge ecosystem; Kinesis is…~8 min
  4. Streaming vs BatchBatch processes on a schedule; streaming reacts per-event.~8 min
  5. Orchestration vs ChoreographyOrchestration vs choreography for coordinating microservices, decided by where control lives: orchestration puts one central coordinato…~9 min
  6. Monolith vs MicroservicesMicroservices don’t remove complexity, they relocate it from the codebase into the network.~9 min

Ship it, then keep it up

Package it, schedule it, and be able to answer "is it broken?" without guessing — plus the review culture that decides how fast you can change it afterwards.

0 / 4 read · 57m
  1. The Docker Handbook Start here A container is a lie told by the kernel — namespaces (what a process sees) plus cgroups (what it uses), not a VM.~12 min
  2. The Kubernetes HandbookThe one idea under all the YAML — declare desired state, and a control loop makes reality match.~15 min
  3. The Observability HandbookSeeing inside production — monitoring vs observability, the three pillars (metrics, logs, traces) and what each answers, structured log…~15 min
  4. The Code Review Culture HandbookThe quantitative case for code review plus the human norms that make it pay off.~15 min

AI engineering

The stack this site exists for — from what a transformer is doing to what an agent costs you in production, in the order the pieces depend on each other.

How models actually work

Fundamentals, then the transformer, then the four architectural forks — encoder or decoder, dense or sparse, autoregressive or diffusion — and finally what people have managed to read off the weights.

0 / 11 read · 2h 37m
  1. ML Fundamentals Start here The seven concept pairs every practitioner is expected to have straight — how machines learn, what they predict, the two ways they miss…~28 min
  2. The Transformers HandbookKey handbookThe architecture behind every LLM, built up from scratch — tokens and embeddings, positional encoding, self-attention (query/key/value)…~14 min
  3. The Embeddings HandbookMeaning as geometry: how text and images become vectors so that similar meaning lands nearby, why cosine similarity measures direction…~15 min
  4. Encoder vs Decoder ModelsEncoder vs decoder Transformers, decided by which way attention flows: an encoder reads the whole input at once (bidirectional) to buil…~9 min
  5. MoE vs Dense ModelsMixture of Experts vs dense LLMs, decided by which parameters fire per token: a dense model runs every parameter on every token; an MoE…~9 min
  6. The Reinforcement Learning HandbookLearning to act by trial and error — the agent-environment loop, cumulative reward and discounting, exploration vs exploitation, value…~15 min
  7. The Diffusion Models HandbookHow AI images are really made — generation as iterative denoising, the forward noising and reverse denoising processes, the elegant noi…~14 min
  8. Diffusion LLMsText generation that isn't one token at a time.~13 min
  9. Autoregressive vs Diffusion LLMsAutoregressive vs diffusion language models, decided by how text is generated: autoregressive models emit one token at a time left-to-r…~9 min
  10. World ModelsWhat it means for AI to learn a predictive model of an environment it can imagine inside — the basis of model-based RL, planning, and c…~15 min
  11. Mechanistic InterpretabilityThe engineer's guide to reading a model — interpretability as a debugger, not a séance.~16 min

Prompting and context

The only ladder here with no prerequisites. What a good prompt is, what belongs in the window at all, how to get a shape back you can parse, and how to stop paying for the same tokens twice.

0 / 4 read · 1h 28m
  1. The Prompting Handbook Start here A friendly, hands-on field guide for everyday humans — learn the CRISP framework, spot bad prompts, practice with real recipes, play a…~45 min
  2. The Context Engineering HandbookKey handbookThe discipline that replaced prompt-tweaking: curating everything the model sees under a finite attention budget.~13 min
  3. The Structured Outputs HandbookHow LLMs return guaranteed-valid JSON for function calling and tool use.~15 min
  4. The Prompt Caching HandbookHow prompt caching actually works.~15 min

Retrieval

RAG, the store underneath it, and the three comparisons you cannot answer until you have read both sides — vectors or keywords, rerank or not, retrieve or just fine-tune.

0 / 7 read · 1h 14m
  1. The RAG Handbook Start here Retrieval-augmented generation end to end — why LLMs need it, chunking, embeddings and vector search, hybrid retrieval, two-stage reran…~15 min
  2. The Vector Databases HandbookHow vector search actually works — why brute-force kNN doesn't scale, distance metrics, approximate nearest neighbor, the HNSW, IVF and…~15 min
  3. Embeddings vs Keyword SearchEmbeddings match meaning even with zero shared words; keyword search (BM25) matches exact terms with a precision embeddings can’t guara…~8 min
  4. Cross-Encoder vs Bi-EncoderCross-encoder vs bi-encoder for semantic search, decided by when the query meets the document: a bi-encoder embeds query and document s…~9 min
  5. SQL vs Vector DatabaseSQL vs vector databases, decided by the kind of question you ask: a relational database answers exact, structured queries over rows (WH…~9 min
  6. RAG vs Long ContextA 1M-token window makes it tempting to skip retrieval entirely — but bigger context doesn’t mean the model reads it evenly, and it isn’…~9 min
  7. RAG vs Fine-TuningKey handbookThe most common wrong turn in AI engineering — reaching for fine-tuning when you needed retrieval, or vice versa.~9 min

Building agents

Patterns first, then the honest question of whether you need an agent at all, then memory, the loop, the harness — and what breaks once the loop runs for hours or spends money.

0 / 10 read · 4h 29m
  1. The Agent Patterns Handbook Start here The design patterns behind every LLM agent — the ReAct Thought–Action–Observation loop, tool/function calling, plan-and-execute, reflec…~14 min
  2. Agents vs WorkflowsAutonomy is a cost, not a default.~9 min
  3. The Agent Memory HandbookA finite context window forces a memory strategy.~15 min
  4. The Loop Engineering HandbookKey handbookStop prompting your agents — design the system that prompts them for you.~18 min
  5. The Harness Engineering HandbookThe loop that wraps a model into an autonomous agent — plan, act, observe — and why its most important property is bounds.~15 min
  6. Long-Horizon AgentsWhy a 10-step task the agent nails becomes a 100-step task it never finishes — reliability COMPOUNDS: a single run of k steps succeeds…~14 min
  7. The Multi-Agent Orchestration HandbookCoordinating multiple LLM agents on one task — and when not to.~15 min
  8. Self-Improving AgentsThe long one: how an agent gets better at its own job, end to end.~140 min
  9. The Browser & Computer-Use Agents HandbookLLM agents that operate a UI like a human — no API required.~15 min
  10. Agentic PaymentsHow to let an agent pay without letting it drain your account.~14 min

Tools, protocols and interop

How an agent reaches the outside world. MCP as the wire, skills as the packaging, A2A as the agent-to-agent case — with each comparison placed after both things it compares.

0 / 7 read · 1h 15m
  1. The Model Context Protocol (MCP) Handbook Start here The open standard connecting AI agents to the outside world — why M×N bespoke integrations collapse to M+N, the host/client/server role…~11 min
  2. MCP vs Function CallingPitted against each other, but they live at different layers — like comparing USB-C to sending data.~8 min
  3. MCP AppsInteractive UIs inside Claude and ChatGPT — the first official UI extension to the Model Context Protocol.~13 min
  4. The Agent Skills HandbookKey handbookPackaged, on-demand expertise for AI agents: what a skill folder actually is (SKILL.md + scripts + resources), progressive disclosure a…~11 min
  5. Agent Skills vs MCPAgent Skills vs the Model Context Protocol, decided by what each adds: a Skill is packaged knowledge and procedure loaded into the agen…~9 min
  6. MCP vs A2ADifferent layers, not rivals: MCP connects an agent to its own tools and data; A2A connects one autonomous agent to another.~8 min
  7. The A2A Agent Interop HandbookLetting heterogeneous agents from different teams work together.~15 min

Training and adaptation

Changing the weights, when prompting has run out. Fine-tuning and its cheap variant, the machinery for training across many GPUs, and the RL route that replaced human preference with a checkable reward.

0 / 7 read · 1h 38m
  1. The Fine-Tuning Handbook Start here When to fine-tune vs RAG vs prompting, full fine-tuning vs parameter-efficient methods, how LoRA and QLoRA make it cheap enough for a s…~15 min
  2. LoRA vs Full Fine-TuningLoRA freezes the base model and trains a tiny low-rank adapter; full fine-tuning updates every weight.~9 min
  3. The Distributed Training HandbookWhen the model doesn't fit, split it four ways.~13 min
  4. The Synthetic Data HandbookUsing LLMs to generate training and eval data.~15 min
  5. Test-Time Compute & Reasoning ModelsKey handbookWhy letting a model think longer at inference — long chain-of-thought, sampling with self-consistency, best-of-N with a verifier, and s…~15 min
  6. RL from Verifiable Rewards (RLVR)The training technique behind modern reasoning models: reinforcement learning where the reward comes from a programmatic check (a unit…~16 min
  7. RL Environments EngineeringThe build side of verifiable rewards — the reasoning models everyone celebrates are taught by the ENVIRONMENT, not the algorithm.~15 min

Serving models, and paying for them

The longest ladder here, because this is where the bill is. Serving engines, the hardware under them, the four levers that cut cost, and the two reference tables to check your arithmetic against.

0 / 13 read · 2h 28m
  1. The LLM Serving Handbook Start here How to serve large language models fast and cheap — prefill vs decode, the KV cache, continuous batching and PagedAttention (vLLM), qua…~15 min
  2. vLLM vs TGI vs SGLangThree open-source LLM serving engines that all do continuous batching but differ in their signature KV-cache trick: vLLM’s PagedAttenti…~9 min
  3. The GPU Fundamentals HandbookThe bottleneck is never where you think.~12 min
  4. GPU vs TPUGPUs are flexible general-purpose processors with the CUDA ecosystem behind them; TPUs are ASICs purpose-built around the systolic arra…~8 min
  5. Quantization vs DistillationQuantization vs distillation for shrinking LLMs, decided by what you change: quantization keeps the same model but stores its weights i…~9 min
  6. Batch vs Real-Time InferenceBatch vs real-time (online) inference, decided by whether a human is waiting: batch processes many inputs together offline for maximum…~9 min
  7. The Model Routing HandbookSend each request to the cheapest model that can handle it.~15 min
  8. LLM vs SLMThe choice is about deployment constraints, not just raw capability: LLMs run in the cloud with broad reasoning; SLMs run on-device wit…~8 min
  9. The AI Cost Engineering HandbookAI spend is a sum over tokens: input-tokens × input-price + output-tokens × output-price, per million.~15 min
  10. LLM API PricingEvery major provider’s API pricing in one comparable table, per million tokens — Anthropic Claude, OpenAI GPT, Google Gemini — pulled f…~10 min
  11. Tokenizer Efficiency BenchmarkAn original, reproducible benchmark of 9 real tokenizers across English prose, code, and non-English text — measuring characters-per-to…~11 min
  12. The Local LLM Stack HandbookRunning LLMs locally with ollama and llama.cpp.~15 min
  13. The On-Device AI HandbookSmall models, everywhere: why 1–14B models got frontier-adjacent (distillation + textbook data), the napkin law of local inference (tok…~12 min

Evals and observability

The part that decides whether any of the above is working. Agent evals, then the same thing wired into CI, then what to log in production — and the interview questions that probe all three.

0 / 4 read · 2h 25m
  1. The Agent Evaluations Handbook Start here A self-contained handbook on evaluating AI agents — theory, interactive widgets, and practical guidance.~60 min
  2. The Evals in CI HandbookTurn LLM quality into a regression gate, like unit tests for code.~15 min
  3. The LLM Observability HandbookTracing an AI request as a tree of spans.~15 min
  4. 51 LLM Evals Interview QuestionsGolden sets, LLM-as-judge, regression testing, offline vs online evals, RAG evals, agent evals, red-teaming, and observability — demyst…~55 min

Security, safety and the law

Attacks first, then the three defences that actually hold — guardrails, a sandbox, and never letting the data in — and finally the regulation that makes some of it mandatory.

0 / 6 read · 1h 29m
  1. The AI Security Handbook Start here Your app now reads the internet and believes it.~14 min
  2. The Guardrails Engineering HandbookThe input and output filters that keep an LLM system safe.~15 min
  3. The Sandboxing HandbookSafely running untrusted or agent-generated code.~15 min
  4. The PII in LLM Pipelines HandbookHandling personal data safely with the redact-before-send pattern: detect PII, replace each value with a stable placeholder before the…~15 min
  5. Agent Supply-Chain SecurityThe other door into your agent — the one you open yourself.~14 min
  6. The EU AI Act HandbookThe world's first comprehensive AI law, made legible for engineers.~16 min

Shipping an AI product

Everything above assembled into something with users: the product surface, the data pipeline feeding it, a real-time voice case, and how the code itself now gets written.

0 / 4 read · 59m
  1. The AI Product Engineering Handbook Start here A demo isn't a product.~16 min
  2. The Data Engineering for AI HandbookThe pipeline that turns raw documents into usable data for RAG and training — ingest, chunk, deduplicate, quality-filter.~15 min
  3. The Voice AI Real-Time Agents HandbookWhy a voice agent is a stopwatch, not a chat box.~15 min
  4. The Agentic Coding HandbookKey handbookWorking with coding agents — Claude Code, Cursor and their cousins — without drowning in AI slop.~13 min

Forward deployed

Engineering inside somebody else’s company: the role, the deployment loop, the compliance wall, and the interviews that gate it.

What the job actually is

Start here if the title is new to you. What an FDE does, what a week looks like, whether it is worth doing — and the two adjacent titles that mean nearly the same thing.

0 / 7 read · 1h 32m
  1. What Is a Forward Deployed Engineer? Start here The pillar definition, with sources: an FDE owns the full arc of a deployment inside the customer’s environment.~16 min
  2. The Forward Deployed Engineer: Role & MindsetThe fastest-growing engineering title in AI, demystified: what an FDE actually does (owns the customer outcome, not the ticket), how it…~14 min
  3. A Day in the Life of an FDEHour by hour, built from the published first-hand accounts: a customer-site day, a remote build day, and a launch-week day — plus how t…~10 min
  4. Is Forward Deployed Engineering Worth It?The honest version.~14 min
  5. Applied AI Engineer, Explained“Applied” does not mean applied research — it means the model meets a real customer’s systems and you are standing where that happens.~13 min
  6. AI Solutions EngineerThe engineer on the other side of the signature.~13 min
  7. The FDE CanonTen sources that actually explain the role — first-hand accounts, job descriptions read as documents, the commercial argument, and the…~12 min

FDE against every adjacent role

Six comparisons and a salary breakdown, read in one sitting, so you can tell which job a posting is actually describing before you apply to it.

0 / 7 read · 1h 18m
  1. FDE vs Every Adjacent Role Start here One master table placing the Forward Deployed Engineer against eleven adjacent roles — software engineer, AI engineer, solutions engine…~9 min
  2. FDE vs Software EngineerSame code, different physics.~12 min
  3. FDE vs AI EngineerIdentical toolbox — RAG, agents, tool schemas, evals — but the AI engineer’s hard problem is the model and the FDE’s is the organisatio…~11 min
  4. FDE vs Solutions Engineer & ArchitectOne question separates all three: after signature, who is accountable for what runs in production?~11 min
  5. FDE vs ConsultantIs forward deployed engineering just consulting with better branding?~10 min
  6. FDE vs FDSE vs Deployment StrategistPalantir split forward-deployed work into an engineering ladder and a strategy ladder; most companies copied the title and not the spli…~12 min
  7. Forward Deployed Engineer Salary, ExplainedWhy the public numbers differ by seven times, reconciled: base versus total comp, six employer tiers from frontier labs to defense inte…~13 min

The deployment loop

The job in the order it happens: scope it, learn the domain, prototype, integrate, then show the customer something that makes the next meeting happen.

0 / 8 read · 1h 44m
  1. Customer Discovery & Scoping for FDEs Start here How FDEs find the real problem before building the wrong one: the wrong-problem tax, watching users instead of asking, the questions th…~13 min
  2. Learning a Customer’s Domain FastHow FDEs get dangerous in a new field in a week: why domain fluency is a superpower, learning from the customer’s real data, running SM…~10 min
  3. Rapid Prototyping for FDEsHow FDEs build a convincing prototype in days, not months: throwaway-first thinking, the smallest demoable vertical slice, choosing a b…~12 min
  4. Customer Integration Patterns for FDEsIntegration is where POCs die.~15 min
  5. AI App Patterns for FDEsThe AI toolkit a modern FDE deploys: RAG over the customer’s documents, agents that drive their workflows, and evals as the acceptance…~14 min
  6. MCP for FDEsAnthropic’s FDE job description names MCP servers as a deliverable.~17 min
  7. The Demo & Storytelling PlaybookA demo is a performance, not a code review.~11 min
  8. Solutions Engineering & Stakeholder ManagementThe non-code half of the FDE job: mapping the room (champion, blocker, decision-maker, user), managing scope creep with a visible later…~12 min

Customer-grade production

What a demo has to survive to become a deployment: their cloud, their auditors, their identity provider, their on-call rota — and finally the handoff that ends your involvement.

0 / 11 read · 2h 50m
  1. Deploying in Customer Environments Start here Ship where the customer can actually run it: the four deployment targets (your SaaS, the customer’s cloud/VPC, on-prem, and air-gapped)…~13 min
  2. IaC for Customer EnvironmentsTerraform and Helm when the account, the cluster and the change process all belong to someone else: who holds state, the seven things a…~17 min
  3. From POC to Production: HardeningThe gap between "it demoed" and "people depend on it": how FDEs harden a proof of concept into a real system — reliability (sad-path ha…~13 min
  4. Incidents, On-Call & SLAs for FDEsProduction in an environment you do not own: negotiating an SLA whose clock starts when you are paged (not when a user noticed), surviv…~16 min
  5. Security & Compliance for FDEsEnough security to earn a customer’s "yes": what SOC 2, HIPAA, GDPR and PCI actually ask, how to handle data and secrets, what a securi…~14 min
  6. Enterprise Identity for FDEsThe layer that blocks more enterprise deals than model quality ever will.~17 min
  7. Air-Gapped & Regulated DeploymentFive things customers mean by “air-gapped”, FedRAMP and DoD impact levels explained for engineers, the eight-step path model weights ta…~18 min
  8. Evals for Client WorkThe most-cited hard skill in frontier-lab FDE job descriptions and the least taught anywhere: turning “it should be accurate” into a gate.~18 min
  9. Proving Value & ROIRoughly 95% of enterprise AI pilots showed no measurable P&L impact — and “measurable” is doing the work in that sentence.~15 min
  10. The Commercial Layer for FDEsThe part nobody teaches: reading a SOW before you are bound by it, the six clauses that cause all the pain, four pricing models and wha…~18 min
  11. Handoff & Enablement for FDEsHow FDEs hand off a deployment without becoming a permanent, unpaid support team: why the exit is the goal, what to hand off (runbook,…~11 min

Getting hired as an FDE

The route in, the nine rounds, the take-home — then five real loops broken down company by company, from Palantir’s FDSE to Distyl’s AI-allowed interview.

0 / 8 read · 1h 34m
  1. How to Become a Forward Deployed Engineer Start here The concrete route in: the real bar (staff-level judgement, not SWE-lite), six skills in dependency order, the one portfolio artefact t…~15 min
  2. FDE Interview Prep: The 9 RoundsThe complete Forward Deployed Engineer interview guide — decomposition, learning round, incremental coding, debugging/re-engineering, t…~17 min
  3. The FDE Take-Home PlaybookTake-homes are the centre of gravity in modern FDE loops and they are graded on judgement, not volume.~12 min
  4. The Palantir FDSE InterviewThe loop that every other company copied: the online assessment, the panel drawn from a pool of five round types, and the decomposition…~12 min
  5. The OpenAI FDE InterviewHow the reported OpenAI forward deployed engineer loop works: the multi-hour API take-home, the required video walkthrough where it is…~10 min
  6. The Anthropic Applied AI / FDE InterviewThe most agent-shaped FDE role at a frontier lab: the escalating-constraint coding assessment, the deployment case where negotiated eva…~10 min
  7. The Sierra Agent Engineer InterviewLive incremental coding, a support agent you must build without an agent framework, and a debugging onsite with planted bugs.~9 min
  8. The Distyl AI Interview (AI Allowed)One of the few companies that publishes its interview philosophy: use AI in the take-home, then defend everything it produced.~9 min

Your career

Getting hired, getting promoted, and what your specific job looks like once models do part of it.

Getting hired, generally

The artefacts before the rounds: a portfolio, then behavioural, then system design, then the two AI-engineering loops — and the negotiation you skip at your own cost.

0 / 8 read · 4h 8m
  1. The Resume & Portfolio Handbook Start here Your resume has two readers and the first isn't human.~15 min
  2. The Behavioral Interview HandbookThe round that decides your level — why behavioral interviews matter and what they signal, the STAR method with a worked example, build…~14 min
  3. 50 System Design Interview QuestionsThe questions that actually come up in senior and staff loops, each with a concise, defensible answer you can say out loud: scaling and…~40 min
  4. The Agentic AI Interview HandbookTwenty topics every senior AI engineer should be able to reason about live — from eval pipelines to reliability patterns for generative…~60 min
  5. The Senior AI Engineer Interview Handbook60 questions across architecture, production incidents, agentic systems, RAG, evals, cost, safety, and leadership — what staff-level AI…~75 min
  6. AI-Assisted InterviewsThe 2026 technical-interview FORMAT (distinct from the content handbooks): AI is now allowed in the room, so the exam shifts from writi…~13 min
  7. The Salary Negotiation HandbookThe expected-value case for always countering an offer.~15 min
  8. The Staff Engineer Behaviors HandbookThe hardest promotion in engineering — senior to staff — trips people up because they aim at the wrong target: writing more/better code…~16 min

Your engineering role in the AI era

One handbook per discipline on what the job is becoming, closing with the two arguments about how the work itself gets done now — vibe or spec, and which CLI.

0 / 10 read · 2h 25m
  1. The New Grad in the AI Era Start here A straight answer to “AI is killing junior roles — what do I even learn?” The junior grind (boilerplate, glue, first-draft tests) is be…~15 min
  2. The AI-Era Backend EngineerHow the backend-engineer job changes in the AI era — and how to be the one who gets more valuable, not less.~16 min
  3. The AI-Era Frontend EngineerHow the frontend-engineer job changes in the AI era — and how to be the one who gets more valuable.~16 min
  4. The AI-Era Data EngineerHow the data-engineer job changes in the AI era.~16 min
  5. The AI-Era DevOps EngineerHow the DevOps / platform-engineer job changes in the AI era.~16 min
  6. The AI-Era QA EngineerHow the QA / test-engineer job changes in the AI era.~16 min
  7. The AI-Era Engineering ManagerHow the engineering-manager job changes in the AI era.~16 min
  8. The AI-Era Product ManagerHow the product-manager job changes in the AI era — and how to be the PM who gets more valuable.~16 min
  9. Vibe Coding vs Spec-Driven DevelopmentVibe coding vs spec-driven development, two ways to build with AI: vibe coding is conversational and exploratory — prompt, run, keep wh…~9 min
  10. Claude Code vs Codex vs Gemini CLIThree terminal-native agentic coding tools from Anthropic, OpenAI and Google that share one loop — read the repo, plan, edit files, run…~9 min

Switching lanes

Four routes in from somewhere else — no tech background, an adjacent hype cycle, independent consulting, or founding the thing yourself.

0 / 4 read · 1h
  1. Career Switch to Tech Start here Switching into tech got harder and easier at the same time, and pretending otherwise helps no one.~15 min
  2. Blockchain to AIA skills-transfer map for blockchain and crypto engineers moving into AI.~15 min
  3. The AI ConsultantEvery company wants to adopt AI and most don't know how — the opening for AI consultants and freelancers.~15 min
  4. AI for FoundersAI collapses the cost of building — one person or a tiny team can now ship what used to take a whole engineering org.~15 min

AI for the rest of the product team

The people an engineer ships alongside. Same structure each time: what the model is genuinely good at in that seat, what it is not, and the workflow to adopt first.

0 / 11 read · 2h 49m
  1. AI for Product Managers Start here The practical middle ground for PMs: where AI genuinely helps (PRD/spec drafts, synthesizing research you actually gathered, competitiv…~18 min
  2. AI for DesignersA practical handbook on using AI in design: where it genuinely helps (exploring variations, first-draft copy and assets, synthesizing r…~16 min
  3. AI for Technical Program ManagersAI writes your status reports, summarizes the meeting, updates the tracker, and drafts the project plan — the reporting layer of progra…~15 min
  4. AI for Scrum MastersAn honest handbook, because this role is under more pressure than most: standups, sprint analytics, backlog grooming, burndown reports,…~15 min
  5. AI for Business AnalystsAI drafts requirements docs, turns meeting notes into user stories, summarizes stakeholder calls, and writes the SQL for your reports.~15 min
  6. AI for Data AnalystsText-to-SQL and auto-dashboards are automating the mechanical half of analysis.~15 min
  7. AI for Technical WritersAI drafts docs, generates API reference from code, and answers user questions directly — so fewer people read the docs at all.~15 min
  8. AI for Developer RelationsAI drafts your blog posts, scaffolds tutorials, writes the sample code, and answers routine developer questions — and, bigger, develope…~15 min
  9. AI for Support EngineersSupport automation is the #1 deployed AI use case, so this role is changing fast.~15 min
  10. AI for Network EngineersAI and automation are eating the manual half of network engineering — per-device CLI config, routine changes, first-pass log analysis,…~15 min
  11. AI for Game DevelopersAI generates concept art, textures, 3D assets, boilerplate gameplay code, dialogue drafts, and procedural content — and it is reshaping…~15 min

AI for your profession

Written for people who do not write code: twelve professions, each with the tasks worth handing over, the ones to keep, and the confidentiality line you cannot cross.

0 / 12 read · 3h 42m
  1. AI for Lawyers Start here The practical middle ground for legal professionals: where AI genuinely helps (first drafts, brief summaries, review triage), the three…~20 min
  2. AI for DoctorsThe practical middle ground for clinicians: where AI genuinely helps (notes from dictation, discharge instructions, prior auths, litera…~20 min
  3. AI for AccountantsThe practical middle ground for accounting and tax professionals: where AI genuinely helps (first drafts, document summaries, reconcili…~18 min
  4. AI for Financial AnalystsThe practical middle ground for financial analysis: where AI genuinely helps (summarizing filings and transcripts, first-draft memos an…~18 min
  5. AI for HRThe practical middle ground for HR and recruiting: where AI genuinely helps (job descriptions, first-draft policies, structured intervi…~18 min
  6. AI for TeachersThe practical middle ground for educators: where AI genuinely helps (lesson plans, differentiating materials into reading levels, first…~18 min
  7. AI for JournalistsThe practical middle ground for reporters and editors: where AI genuinely helps (transcription, document/report summaries, headline and…~18 min
  8. AI for MarketersThe practical middle ground for marketers: where AI genuinely helps (first-draft copy, ideation, repurposing one asset into many format…~18 min
  9. AI for SalesThe practical middle ground for sales reps: where AI genuinely helps (personalized outreach from real research, follow-ups, call summar…~18 min
  10. AI for ArchitectsThe practical middle ground for architectural practice: where AI genuinely helps (concept ideation, first-draft narratives and specs, d…~18 min
  11. AI for Civil EngineersThe practical middle ground for civil and structural engineers: where AI genuinely helps (observation reports, RFI responses, meeting m…~20 min
  12. AI for Real EstateThe practical middle ground for agents and brokers: where AI genuinely helps (listing descriptions, client follow-ups, market-data summ…~18 min

Handbooks — frequently asked questions

Are the handbooks free to read?

Yes. Every handbook is free, full-length, and self-contained — no paywall, no sign-up, and no email required to read any of them.

Which handbook should I read first?

The Prompting Handbook — rung 1 of the Prompting and context ladder, and the only handbook here that assumes nothing before it. If you are an engineer looking for the technical floor instead, start at ML Fundamentals (rung 1 of "How models actually work") or System Design Fundamentals (rung 1 of the distributed-systems ladder).

Why are the handbooks in ladders instead of categories?

Because reading order changes what a handbook teaches. Each rung is written assuming the one below it, so a ladder can tell you what to read next; a category list cannot. It also puts every "X vs Y" comparison directly after both things it compares — RAG vs Fine-Tuning is the last rung of the retrieval ladder, because it is only answerable once you have read both sides.

Who writes these handbooks?

They are written by Saurabh Singh, a senior AI engineer, based on real production experience building agentic AI systems, evals, and large-scale services — not generic content.

How long does each handbook take to read?

Most run between 8 and 20 minutes, shown on each rung and filterable with the length pills. A handful of the older flagship handbooks are much longer — the Prompting, Senior AI Engineer and System Design Fundamentals handbooks are designed as a single focused sitting, with interactive widgets and quizzes throughout.

Do the handbooks track what I have read?

Yes, in your own browser. Marking a handbook complete fills in its rung and its ladder’s progress bar, and the catalog surfaces the next rung of whichever ladder you are furthest along. Nothing is sent anywhere and there is no account.

Request · Open Channel

Something missing?

If there's a topic you'd like covered — or a rung you think sits in the wrong place on one of the ladders above — send it through. Genuine requests shape what gets written next.