Meta tags:
description= Learn it. Build it. Ship it for others. Contribute to rohitg00/ai-engineering-from-scratch development by creating an account on GitHub.;
Headings (most frequently used words):
the, uh, oh, of, agent, code, navigation, your, this, lesson, every, saved, searches, from, to, files, history, memory, works, shape, contributing, sponsor, license, footer, course, into, search, repositories, users, issues, pull, requests, provide, feedback, rohitg00, ai, engineering, scratch, menu, use, filter, results, more, quickly, folders, and, latest, commit, repository, creator, persistent, which, naturally, with, any, agents, or, chat, assistants, how, curriculum, getting, started, ships, something, contents, toolkit, where, start, why, matters, now, work, star, about, releases, project, packages, contributors, languages, prerequisites, built, in, skills, claude, cursor, codex, openclaw, hermes, fig_002, worked, sample, phase, setup, tooling, 12, lessons, install, skill, drop, workbench, own, repo, browse, entire, as, json, smoke, check, python, topics, resources, conduct, stars, watchers, forks,
Text of the page (most frequently used words):
python (508), build (304), learn (138), and (117), the (110), phase (80), agent (78), lesson (49), from (41), end (40), for (36), scripts (33), you (32), #scratch (31), llm (31), agents (31), with (31), code (30), skills (30), your (28), lessons (27), python3 (25), engineering (23), mcp (22), type (22), every (21), harness (21), lang (21), github (20), any (20), multi (20), learning (19), vision (19), models (19), use (17), p11 (17), p17 (17), production (17), nlp (16), repo (16), skill (16), rag (16), evaluation (16), json (15), run (15), train (15), language (14), loop (14), multimodal (14), research (14), audio (14), that (13), this (13), security (13), deep (13), attention (13), claude (13), target (13), prompt (13), safety (13), p14 (13), one (12), model (12), path (12), tuning (12), self (12), p13 (12), sub (12), not (11), only (11), name (11), all (11), workbench (11), phases (11), outputs (11), advanced (11), tool (11), eval (11), p18 (11), video (11), com (10), curriculum (10), optimization (10), diffusion (10), install_skills (10), framework (10), open (10), time (9), readme (9), transformers (9), what (9), llms (9), why (9), into (9), real (9), distributed (9), alignment (9), auto (9), p15 (9), memory (9), generation (9), julia (8), loading (8), reload (8), page (8), sponsors (8), rohitg00 (8), ship (8), data (8), text (8), context (8), math (8), each (8), tools (8), fine (8), p10 (8), search (8), typescript (7), there (7), contributing (7), full (7), hours (7), file (7), capstone (7), cursor (7), openclaw (7), layout (7), training (7), pipeline (7), cross (7), vlm (7), token (7), autonomous (7), inference (7), understanding (7), voice (7), architecture (7), speech (7), view (7), can (6), docs (6), 2026 (6), rust (6), license (6), star (6), history (6), directory (6), check (6), setup (6), skip (6), first (6), lesson_run (6), same (6), pack (6), codex (6), hermes (6), files (6), prompts (6), retrieval (6), linear (6), function (6), call (6), decoding (6), level (6), projects (6), fundamentals (6), modeling (6), image (6), enterprise (6), community (5), navigation (5), was (5), error (5), while (5), please (5), more (5), project (5), mit (5), computer (5), non (5), read (5), direct (5), rlhf (5), new (5), know (5), core (5), foundations (5), default (5), actually (5), commit (5), build_catalog (5), task (5), source (5), dry (5), ships (5), reusable (5), add (5), install (5), coding (5), constitutional (5), parallel (5), state (5), metrics (5), query (5), encoder (5), embeddings (5), transformer (5), window (5), genai (5), p12 (5), speculative (5), systems (5), openai (5), frameworks (5), selection (5), features (5), caching (5), benchmarks (5), classification (5), long (5), session (5), feedback (5), building (5), understand (5), actions (5), loss (5), support (5), built (5), other (4), sponsor (4), stars (4), conduct (4), resources (4), generative (4), swarm (4), course (4), machine (4), want (4), sign (4), block (4), get (4), 503 (4), rule (4), schema (4), quiz (4), audit_lessons (4), glossary (4), team (4), anthropic (4), protocol (4), preference (4), gpt (4), are (4), industry (4), start (4), head (4), runs (4), execute (4), per (4), under (4), result (4), workflow (4), catalog (4), tooling (4), writes (4), artifact (4), scaffold_workbench (4), overwrite (4), own (4), reads (4), npx (4), them (4), injection (4), critic (4), gradient (4), instruction (4), observability (4), server (4), p16 (4), document (4), semantic (4), native (4), llama (4), apis (4), stack (4), platform (4), world (4), patterns (4), improvement (4), calling (4), refresh (4), streaming (4), vit (4), networks (4), detection (4), neural (4), intuition (4), msg (4), fig_001 (4), find (4), 3553ff (4), issues (4), action (3), information (3), contributors (3), main (3), report (3), intelligence (3), reinforcement (3), topics (3), aiengineeringfromscratch (3), fork (3), rate (3), rules (3), via (3), logo (3), medium (3), feature (3), coderabbit (3), free (3), work (3), fails (3), shape (3), keys (3), issue (3), inside (3), output (3), before (3), code_of_conduct (3), roadmap (3), track (3), lesson_template (3), forking (3), shot (3), software (3), programming (3), where (3), list (3), automated (3), deps (3), script (3), stdout (3), syntax (3), mode (3), execution (3), most (3), after (3), existing (3), does (3), site (3), verify (3), have (3), top (3), hand (3), minimal (3), force (3), preview (3), init (3), flat (3), tag (3), cloning (3), clone (3), produces (3), yaml (3), content (3), runner (3), decomposition (3), hybrid (3), chunking (3), pretraining (3), layer (3), bpe (3), tokenizer (3), verification (3), plan (3), flow (3), registry (3), complete (3), red (3), devops (3), assistant (3), guard (3), metr (3), stable (3), differential (3), visual (3), family (3), reward (3), serving (3), vllm (3), economics (3), testing (3), cost (3), primitive (3), eagle (3), coordination (3), modes (3), scaling (3), theory (3), debate (3), protocols (3), swarms (3), scale (3), then (3), durable (3), vector (3), graph (3), sampling (3), structured (3), dive (3), whisper (3), prediction (3), weight (3), clip (3), see (3), app (3), methods (3), gans (3), recognition (3), tracking (3), knowledge (3), segmentation (3), pytorch (3), functions (3), algebra (3), algorithm (3), git (3), its (3), yourself (3), paste (3), works (3), write (3), option (3), p19 (3), already (3), but (3), last (3), changelog (3), insights (3), pull (3), requests (3), signed (3), settings (3), another (3), tab (3), saved (3), documentation (3), grade (3), copilot (3), solutions (3), explore (3), personal (2), manage (2), privacy (2), terms (2), footer (2), languages (2), about (2), https (2), releases (2), repository (2), forks (2), activity (2), others (2), maintained (2), attribution (2), required (2), pricing (2), reach (2), above (2), integration (2), 000 (2), linkedin (2), small (2), supported (2), api (2), tweet (2), views (2), twitter (2), exit (2), zero (2), when (2), answer (2), single (2), fix (2), react (2), reasoning (2), chain (2), thought (2), few (2), need (2), consensus (2), will (2), keep (2), getting (2), better (2), creator (2), our (2), fig_003 (2), signal (2), matters (2), stdlib (2), timeout (2), listing (2), reason (2), wired (2), requires (2), heavy (2), editing (2), drift (2), workflows (2), derived (2), counts (2), disk (2), builds (2), out (2), entire (2), edit (2), contract (2), lives (2), handoff (2), drop (2), nested (2), pick (2), filter (2), folder (2), recommended (2), plug (2), homework (2), classifier (2), taxonomy (2), checkpoint (2), resume (2), analysis (2), collective (2), metric (2), classical (2), system (2), precision (2), strategies (2), fusion (2), modality (2), patches (2), demo (2), iteration (2), scheduler (2), paper (2), experiment (2), generator (2), fsdp (2), warmup (2), tokenized (2), corpus (2), large (2), supervised (2), positional (2), dataset (2), sliding (2), tasks (2), gates (2), control (2), over (2), validation (2), combines (2), series (2), tracks (2), range (2), governance (2), chatbot (2), kubernetes (2), scientist (2), asr (2), tts (2), terminal (2), dual (2), risk (2), ecosystem (2), watermarking (2), group (2), bias (2), frontier (2), fsf (2), capability (2), indirect (2), art (2), persistent (2), helps (2), ethics (2), sglang (2), canary (2), deployment (2), gateways (2), prefill (2), routing (2), reduction (2), edge (2), cache (2), quantization (2), awq (2), gptq (2), gguf (2), fp8 (2), nvfp4 (2), gpu (2), modal (2), infrastructure (2), case (2), failure (2), checkpoints (2), mind (2), emergent (2), a2a (2), handoffs (2), orchestration (2), chat (2), speaker (2), role (2), hierarchical (2), supervisor (2), external (2), policy (2), input (2), swe (2), bench (2), evolutionary (2), chatbots (2), horizon (2), initialization (2), still (2), development (2), runtimes (2), collaboration (2), opentelemetry (2), sdk (2), guardrails (2), based (2), flows (2), actor (2), langgraph (2), planning (2), tree (2), principles (2), async (2), client (2), interface (2), temporal (2), autoregressive (2), next (2), dynamic (2), llava (2), resolution (2), patch (2), contrastive (2), hear (2), across (2), machines (2), lora (2), activation (2), deepseek (2), sparse (2), ppo (2), pre (2), mini (2), pipelines (2), wordpiece (2), sentencepiece (2), transfer (2), gradients (2), rectified (2), create (2), decoder (2), rnns (2), everything (2), processing (2), entity (2), subword (2), tokenization (2), topic (2), question (2), sequence (2), cnns (2), object (2), estimation (2), vocabulary (2), pixels (2), debugging (2), introduction (2), stability (2), entropy (2), started (2), bayes (2), curve (2), regression (2), operations (2), vectors (2), through (2), shell (2), docker (2), environments (2), dev (2), environment (2), follows (2), twenty (2), expand (2), contents (2), debugger (2), step (2), explain (2), implement (2), tool_calls (2), because (2), wrote (2), servers (2), here (2), something (2), questions (2), review (2), ten (2), maps (2), starting (2), personalized (2), hour (2), estimates (2), how (2), just (2), ahead (2), theme (2), base (2), themevariables (2), primarycolor (2), fafaf5 (2), primarytextcolor (2), 1a1a1a (2), primarybordercolor (2), linecolor (2), fontfamily (2), jetbrains (2), mono (2), fontsize (2), flowchart (2), line (2), problem (2), raw (2), thing (2), spine (2), doing (2), don (2), pieces (2), vercel (2), requirements (2), txt (2), gitignore (2), web (2), assets (2), 636 (2), commits (2), message (2), menu (2), quality (2), appearance (2), cancel (2), available (2), searches (2), repositories (2), business (2), developer (2), customer (2), services (2), perform, share, cookies, contact, status, inc, html, javascript, packages, blob, published, watching, 239, watchers, tutorial, suggest, ghumare64, rohit, ghumare, however, teach, sell, appreciated, manual, helped, keeps, alive, card, hard, anchors, hero, fold, dedicated, max, partner, platinum, quarterly, gold, listed, provider, 750, silver, row, launch, day, 250, bronze, backers, backer, tier, iii, current, 593, monthly, visitors, 709, acquisition, channel, verified, licensed, sponsorship, alone, cash, l001, l010, validate, presence, emptiness, rejects, legacy, caused, 102, relative, links, choices, friendly, submitting, invariant, progress, template, school, contribute, goal, acting, prompting, instructgpt, denoising, probabilistic, learners, vaswani, 2017, compounds, knowing, boris, cherny, being, remade, front, eyes, andrej, karpathy, hottest, english, foundational, papers, covered, now, senior, engineer, 100, 200, 270, 306, estimated, background, set, override, domains, aggressively, instagram, link_check_skip, domain1, domain2, second, whose, entry, starts, comment, skipped, opt, needs, pkg1, pkg2, strict, 10s, whole, byte, compiles, catches, regressions, introduce, often, bad, indentation, broken, strings, stray, edits, smoke, rebuilds, committed, stale, reject, warn, yml, filesystem, always, match, downstream, drifted, documented, touch, walks, root, truth, browse, seven, surfaces, starter, fresh, init_agent, schema_version, agent_state, task_board, seeds, schemas, scaffold, convention, written, refuses, destination, exits, colliding, collisions, inventory, grouped, manifest, examples, config, without, writing, explicit, useful, filters, offline, custom, whichever, picks, bundle, aware, command, detects, automatically, 388, templates, toolkit, gate, engine, refusal, detector, jailbreak, sharded, atomic, bubble, optimizer, sharding, ddp, ops, leaderboard, aggregation, perplexity, calibration, exec, spec, format, recall, mrr, ndcg, faithfulness, relevance, rewriting, hyde, reranker, bm25, dense, compared, projection, writer, evaluator, literature, hypothesis, save, accumulation, clipping, mixed, cosine, hdf5, downloader, swap, pretrained, weights, assembly, otel, spans, prometheus, fixture, sandbox, denylist, jail, observation, budget, dispatcher, rpc, newline, delimited, stdio, subsystem, tutor, adaptive, scene, dashboard, migration, upgrade, regulated, vertical, troubleshooting, class, codebase, products, cyber, bio, chem, nuclear, moderation, perspective, mats, redwood, apollo, provenance, cards, echoleak, cves, regulatory, korea, synthid, signature, c2pa, fairness, criteria, individual, counterfactual, representational, harm, welfare, rsp, wmdp, garak, pyrit, ascii, jailbreaks, many, jailbreaking, teaming, pair, attacks, scalable, oversight, weak, strong, despite, subversion, faking, scheming, sleeper, deception, mesa, deceptive, rlaif, sycophancy, amplification, hacking, goodhart, law, following, humanity, optional, hosted, cpp, ollama, tgi, finops, unit, tenant, compliance, soc, hipaa, gdpr, act, iso, 42001, secrets, pii, scrubbing, audit, logs, chaos, sre, incident, response, load, llmperf, perf, growthbook, statsig, shadow, progressive, litellm, portkey, kong, bifrost, lmcache, offloading, disaggregated, decode, nvidia, dynamo, batch, discount, standard, ane, hexagon, webgpu, jetson, region, locality, cold, mitigation, serverless, ttft, tpot, itl, goodput, p99, tensorrt, blackwell, radixattention, prefix, workloads, internals, pagedattention, continuous, batching, chunked, autoscaling, karpenter, kai, fireworks, together, baseten, managed, platforms, bedrock, azure, vertex, studies, mast, groupthink, monoculture, queues, durability, economies, incentives, reputation, marl, maddpg, qmix, mappo, pso, aco, simulation, negotiation, bargaining, voting, consistency, topology, byzantine, fault, tolerance, shared, blackboard, routines, stateless, networked, architectures, specialization, planner, executor, verifier, society, orchestrator, worker, pattern, communication, fipa, acl, heritage, acts, emergence, cais, caisi, societal, horizons, preparedness, deepmind, responsible, overrides, rollback, hitl, propose, kill, switches, circuit, breakers, tokens, budgets, caps, governors, running, browser, permission, landscape, codeact, bounded, designs, recursive, aar, workshop, darwin, gödel, modifying, alphaevolve, quiet, taught, briefing, opens, mission, reviewer, separate, builder, marker, runtime, loops, scope, contracts, boundaries, instructions, executable, constraints, capable, fail, driven, queue, event, cron, pve, defense, break, langfuse, phoenix, opik, conventions, pipecat, livekit, cua, gemini, webarena, osworld, gaia, agentbench, agno, mastra, subagents, store, tracing, crewai, crews, autogen, stateful, graphs, htn, libraries, lifelong, voyager, mem0, blocks, sleep, compute, virtual, memgpt, refine, thoughts, lats, reflexion, verbal, rewoo, sdks, auth, enrollment, jwks, audience, pinning, registries, oauth, poisoning, apps, roots, elicitation, transports, design, calls, interfaces, between, colpali, diagram, embodied, vlas, openvla, gr00t, omni, thinker, talker, af3, million, grounding, mio, janus, pro, decoupled, encoders, show, discrete, unified, transfusion, emu3, chameleon, early, internvl3, qwen, fps, onevision, recipes, naflex, flamingo, gated, blip, former, bridge, modalities, tradeoffs, limiting, qlora, reranking, augmented, representations, cot, techniques, put, checkpointing, recomputation, hogwild, jamba, ssm, walkthrough, dualpipe, parallelism, mtp, nsa, walkthroughs, int8, evals, dpo, sft, deepspeed, 124m, tokenizers, games, sim, a2c, a3c, reinforce, dqn, sarsa, monte, carlo, mdps, states, rewards, foundation, game, playing, var, fid, score, matching, inpainting, outpainting, controlnet, conditioning, latent, ddpm, stylegan, conditional, pix2pix, discriminator, autoencoders, vae, images, draft, repeat, variants, laws, flash, mixture, experts, moe, bart, causal, bert, masked, encoding, sinusoidal, rope, alibi, problems, changed, wer, mos, mmau, leaderboards, anti, spoofing, moshi, hibiki, turn, taking, codecs, encodec, snac, mimi, dac, music, conversion, spectrograms, mel, waveforms, fft, speak, dialogue, niah, ruler, longbench, mrcr, ragas, deepeval, relation, extraction, construction, linking, disambiguation, coreference, embedding, nli, textual, entailment, constrained, unigram, multilingual, lda, bertopic, answering, summarization, translation, mechanism, breakthrough, pos, tagging, syntactic, parsing, named, ner, sentiment, glove, fasttext, word, word2vec, bag, words, idf, representation, stemming, lemmatization, monocular, depth, geometry, mlp, sam, gaussian, splatting, keypoint, pose, ocr, simclr, dino, mae, point, clouds, nerfs, instance, mask, cnn, net, yolo, lenet, resnet, convolutions, channels, color, spaces, vlms, jax, schedules, regularization, dropout, decay, batchnorm, optimizers, sgd, momentum, adam, adamw, mse, relu, sigmoid, gelu, backpropagation, forward, pass, perceptron, until, handling, imbalanced, anomaly, naive, hyperparameter, ensemble, boosting, bagging, stacking, variance, unsupervised, means, dbscan, knn, distance, decision, trees, random, forests, logistic, backbone, stochastic, processes, fourier, transform, complex, numbers, convex, statistics, norms, distances, numerical, tensor, singular, value, dimensionality, pca, sne, umap, divergence, descent, theorem, statistical, thinking, probability, distributions, automatic, differentiation, calculus, derivatives, matrix, transformations, eigenvalues, matrices, behind, profiling, linux, management, editor, jupyter, notebooks, cloud, ready, click, given, trace, identify, went, wrong, debug, style, description, steplimitexceeded, raise, return, continue, tool_result, append, args, max_steps, user, def, agent_loop, 120, lines, pure, dependencies, fig_002, worked, sample, lot, portfolio, artifacts, compatible, deploy, workers, expert, help, narrow, curricula, ends, daily, congratulations, learned, eight, specific, placement, prerequisites, selector, intelligently, installed, completed, three, ways, 13px, motto, idea, concrete, pain, concept, diagrams, sklearn, six, beats, split, library, smaller, version, runnable, implementations, narrative, structure, 12px, floor, roof, lower, layers, wonder, breaking, derive, test, five, minute, videos, copy, deploys, holding, laptop, four, gets, backprop, shows, hood, material, teaches, scattered, post, flashy, somewhere, else, rarely, hook, say, readers, days, 241, 669, 150, 639, 320, closes, gap, students, feel, prepared, professionally, which, naturally, assistants, items, latest, date, folders, tags, branches, additional, options, must, change, notification, notifications, public, dismiss, alert, switched, accounts, resetting, focus, qualifiers, results, quickly, submit, include, email, address, contacted, piece, take, very, seriously, provide, tips, clear, users, jump, premium, ons, powered, collections, trending, archive, program, accelerator, maintainer, lab, programs, fund, developers, partners, trust, center, forum, ebooks, reports, events, webinars, stories, industries, government, manufacturing, financial, healthcare, cases, devsecops, modernization, nonprofits, startups, teams, enterprises, company, size, marketplace, blog, stop, leaks, they, secret, protection, secure, vulnerabilities, application, changes, instant, codespaces, automate, integrate, merge, creation, toggle,
Text of the page (random words):
9 learn python 09 production quantization awq gptq gguf fp8 nvfp4 learn python 10 cold start mitigation for serverless llms learn python 11 multi region llm serving and kv cache locality learn python 12 edge inference ane hexagon webgpu jetson learn python 13 llm observability stack selection learn python 14 prompt caching and semantic caching economics learn python 15 batch apis the 50 discount as industry standard learn python 16 model routing as a cost reduction primitive learn python 17 disaggregated prefill decode nvidia dynamo and llm d learn python 18 vllm production stack with lmcache kv offloading learn python 19 ai gateways litellm portkey kong bifrost learn python 20 shadow canary and progressive deployment learn python 21 a b testing llm features growthbook and statsig learn python 22 load testing llm apis k6 llmperf genai perf build python 23 sre for ai multi agent incident response learn python 24 chaos engineering for llm production learn python 25 security secrets pii scrubbing audit logs learn python 26 compliance soc 2 hipaa gdpr eu ai act iso 42001 learn python 27 finops for llms unit economics and multi tenant attribution learn python 28 self hosted serving selection llama cpp ollama tgi vllm sglang learn python phase 18 ethics safety alignment 30 lessons build ai that helps humanity not optional lesson type lang 01 instruction following as alignment signal learn python 02 reward hacking goodhart s law learn python 03 direct preference optimization family learn python 04 sycophancy as rlhf amplification learn python 05 constitutional ai rlaif learn python 06 mesa optimization deceptive alignment learn python 07 sleeper agents persistent deception learn python 08 in context scheming in frontier models learn python 09 alignment faking learn python 10 ai control safety despite subversion learn python 11 scalable oversight weak to strong learn python 12 red teaming pair automated attacks build python 13 many shot jailbreaking learn python 14 ascii art visual jailbreaks build python 15 indirect prompt injection build python 16 red team tooling garak llama guard pyrit build python 17 wmdp dual use capability evaluation learn python 18 frontier safety frameworks rsp pf fsf learn python 19 model welfare research learn python 20 bias representational harm build python 21 fairness criteria group individual counterfactual learn python 22 differential privacy for llms build python 23 watermarking synthid stable signature c2pa build python 24 regulatory frameworks eu us uk korea learn python 25 echoleak cves for ai learn python 26 model system dataset cards build python 27 data provenance training data governance learn python 28 alignment research ecosystem mats redwood apollo metr learn python 29 moderation systems openai perspective llama guard build python 30 dual use risk cyber bio chem nuclear learn python phase 19 capstone projects 85 lessons 17 end to end products 9 deep build tracks 20 40 hours per project 4 12 lessons per track project combines lang 01 terminal native coding agent p0 p5 p7 p10 p11 p13 p14 p15 p17 p18 python 02 rag over codebase cross repo semantic search p5 p7 p11 p13 p17 python 03 real time voice assistant asr llm tts p6 p7 p11 p13 p14 p17 python 04 multimodal document qa vision first p4 p5 p7 p11 p12 p17 python 05 autonomous research agent ai scientist class p0 p2 p3 p7 p10 p14 p15 p16 p18 python 06 devops troubleshooting agent for kubernetes p11 p13 p14 p15 p17 p18 python 07 end to end fine tuning pipeline p2 p3 p7 p10 p11 p17 p18 python 08 production rag chatbot regulated vertical p5 p7 p11 p12 p17 p18 python 09 code migration agent repo level upgrade p5 p7 p11 p13 p14 p15 p17 python 10 multi agent software engineering team p11 p13 p14 p15 p16 p17 python 11 llm observability eval dashboard p11 p13 p17 p18 python 12 video understanding pipeline scene qa p4 p6 p7 p11 p12 p17 python 13 mcp server with registry and governance p11 p13 p14 p17 p18 python 14 speculative decoding inference server p3 p7 p10 p17 python 15 constitutional safety harness red team range p10 p11 p13 p14 p18 python 16 github issue to pr autonomous agent p11 p13 p14 p15 p17 python 17 personal ai tutor adaptive multimodal p5 p6 p11 p12 p14 p17 p18 python deep build tracks multi lesson series that build a complete subsystem from scratch project combines lang 20 agent harness loop contract a agent harness python 21 tool registry with schema validation a agent harness python 22 json rpc 2 0 over newline delimited stdio a agent harness python 23 function call dispatcher a agent harness python 24 plan execute control flow a agent harness python 25 verification gates and observation budget a agent harness python 26 sandbox runner with denylist and path jail a agent harness python 27 eval harness with fixture tasks a agent harness python 28 observability with otel genai spans and prometheus metrics a agent harness python 29 end to end coding agent on the harness a agent harness python 30 bpe tokenizer from scratch b nlp llm python 31 tokenized dataset with sliding window b nlp llm python 32 token and positional embeddings b nlp llm python 33 multi head self attention b nlp llm python 34 transformer block from scratch b nlp llm python 35 gpt model assembly b nlp llm python 36 training loop and evaluation b nlp llm python 37 loading pretrained weights b nlp llm python 38 classifier fine tuning by head swap b nlp llm python 39 instruction tuning by supervised fine tuning b nlp llm python 40 direct preference optimization from scratch b nlp llm python 41 full evaluation pipeline b nlp llm python 42 large corpus downloader c train end to end python 43 hdf5 tokenized corpus c train end to end python 44 cosine lr with linear warmup c train end to end python 45 gradient clipping and mixed precision c train end to end python 46 gradient accumulation c train end to end python 47 checkpoint save and resume c train end to end python 48 distributed data parallel and fsdp from scratch c train end to end python 49 language model evaluation harness c train end to end python 50 hypothesis generator d auto research python 51 literature retrieval d auto research python 52 experiment runner d auto research python 53 result evaluator d auto research python 54 paper writer d auto research python 55 critic loop d auto research python 56 iteration scheduler d auto research python 57 end to end research demo d auto research python 58 vision encoder patches e multimodal vlm python 59 vision transformer encoder e multimodal vlm python 60 projection layer for modality alignment e multimodal vlm python 61 cross attention fusion e multimodal vlm python 62 vision language pretraining e multimodal vlm python 63 multimodal evaluation e multimodal vlm python 64 chunking strategies compared f advanced rag python 65 hybrid retrieval with bm25 and dense embeddings f advanced rag python 66 cross encoder reranker f advanced rag python 67 query rewriting hyde multi query and decomposition f advanced rag python 68 rag evaluation precision recall mrr ndcg faithfulness answer relevance f advanced rag python 69 end to end rag system f advanced rag python 70 task spec format g eval framework python 71 classical metrics g eval framework python 72 code exec metric g eval framework python 73 perplexity and calibration g eval framework python 74 leaderboard aggregation g eval framework python 75 end to end eval runner g eval framework python 76 collective ops from scratch h distributed train python 77 data parallel ddp from scratch h distributed train python 78 zero optimizer state sharding h distributed train python 79 pipeline parallel and bubble analysis h distributed train python 80 sharded checkpoint and atomic resume h distributed train python 81 end to end distributed training h distributed train python 82 jailbreak taxonomy i safety harness python 83 prompt injection detector i safety harness python 84 refusal evaluation i safety harness python 85 content classifier integration i safety harness python 86 constitutional rules engine i safety harness python yaml 87 end to end safety gate i safety harness python the toolkit every lesson produces a reusable artifact by the end you have outputs prompts prompt templates for every ai task skills skill md files for ai coding agents install them with npx skills add plug them into claude cursor codex openclaw hermes or any agent that reads a skill md agents md directory real tools not homework install every course skill into your agent the repo ships 388 skills and 99 prompts under phases outputs recommended install via skills sh no clone no python detects your agent s skills directory automatically npx skills add rohitg00 ai engineering from scratch every skill npx skills add rohitg00 ai engineering from scratch skill agent loop one skill npx skills add rohitg00 ai engineering from scratch phase 14 one phase skills writes to whichever directory your agent picks up claude skills cursor skills codex skills openclaw s skills folder hermes s bundle path or any skill md aware tool one command every agent advanced offline custom layout via scripts install_skills py requires cloning the repo useful when you need tag filters dry runs or a non default layout python3 scripts install_skills py target every skill default layout skills nested python3 scripts install_skills py target layout skills same as above explicit python3 scripts install_skills py target type all skills prompts agents python3 scripts install_skills py target phase 14 one phase only python3 scripts install_skills py target tag rag filter by tag python3 scripts install_skills py target layout flat flat files python3 scripts install_skills py target dry run preview without writing python3 scripts install_skills py target force overwrite existing files target is the skills directory for your agent examples claude skills cursor skills config openclaw skills skills or any path your agent reads by default the script refuses to overwrite an existing destination and exits with code 1 after listing every colliding path use dry run to preview collisions or force to overwrite every non dry run run writes a manifest json in the target with the full inventory grouped by type and phase pick the layout your agent reads layout path written skills target name skill md nested convention supported by claude cursor codex openclaw hermes by phase target phase nn name md flat target name md drop the agent workbench into your own repo the phase 14 capstone ships a reusable agent workbench pack agents md schemas init verify handoff scripts scaffold it into any repo with python3 scripts scaffold_workbench py path to your repo full pack seeds python3 scripts scaffold_workbench py path to your repo minimal skip docs python3 scripts scaffold_workbench py path to your repo dry run preview only python3 scripts scaffold_workbench py path to your repo force overwrite you get the seven workbench surfaces wired up a starter task_board json and a fresh agent_state json at schema_version 1 from there edit the task edit agents md run scripts init_agent py hand the contract to your agent the pack source lives at phases 14 agent engineering 42 agent workbench capstone outputs agent workbench pack browse the entire course as json scripts build_catalog py walks every phase every lesson every artifact on disk and writes catalog json at the repo root one file every course truth python3 scripts build_catalog py writes repo catalog json python3 scripts build_catalog py stdout to stdout do not touch repo python3 scripts build_catalog py out path to file json the catalog is filesystem derived not readme derived so counts always match what is actually on disk use it for site builds downstream tooling or to verify the readme counts have not drifted schema is documented at the top of the script a github action github workflows curriculum yml rebuilds catalog json on every pr and fails the build if the committed file is stale after editing any lesson run python3 scripts build_catalog py and commit the result or ci will reject the pr the same workflow runs audit_lessons py in warn only mode so existing drift does not block contributors smoke check every lesson s python code scripts lesson_run py byte compiles every py file under each lesson s code directory default mode is syntax check only no execution no api keys no heavy ml deps required catches the regressions contributors introduce most often bad indentation broken f strings stray edits python3 scripts lesson_run py syntax check the whole curriculum python3 scripts lesson_run py phase 14 one phase only python3 scripts lesson_run py json json report on stdout python3 scripts lesson_run py strict exit 1 if any lesson fails python3 scripts lesson_run py execute actually run 10s timeout per lesson execute runs each lesson s code main py or the first py file with a 10 second timeout lessons whose entry file starts with a requires pkg1 pkg2 comment listing non stdlib deps are skipped with reason needs deps the script is opt in and not wired into ci stdlib only python 3 10 set link_check_skip domain1 domain2 to override the default skip list twitter com x com linkedin com instagram com medium com domains that aggressively block automated head get where to start background start at estimated time new to programming and ai phase 0 setup 306 hours know python new to ml phase 1 math foundations 270 hours know ml new to deep learning phase 3 deep learning core 200 hours know deep learning want llms and agents phase 10 llms from scratch 100 hours senior engineer only want agent engineering phase 14 agent engineering 60 hours why this matters now fig_003 a the industry signal fig_003 b foundational papers covered the hottest new programming language is english andrej karpathy tweet software engineering is being remade in front of our eyes boris cherny creator of claude code models will keep getting better the skill that compounds is knowing what to build industry consensus 2026 attention is all you need vaswani et al 2017 phase 7 language models are few shot learners gpt 3 phase 10 denoising diffusion probabilistic models phase 8 instructgpt rlhf phase 10 direct preference optimization phase 10 chain of thought prompting phase 11 react reasoning acting in llms phase 14 model context protocol anthropic phase 13 contributing goal read contribute a lesson or fix contributing md fork for your team or school forking md lesson template lesson_template md track progress roadmap md glossary glossary terms md code of conduct code_of_conduct md before submitting a lesson run the invariant check python3 scripts audit_lessons py full curriculum python3 scripts audit_lessons py phase 14 single phase python3 scripts audit_lessons py json ci friendly output exit code is non zero when any rule fails rules l001 l010 validate directory shape docs en md presence h1 code non emptiness quiz json schema rejects the legacy q choices answer keys that caused issue 102 and relative links insi...
|
