# Himanshu

> Software engineer and former founder. Co-founded AuraX, a diffusion-based virtual
> try-on startup incubated at IIIT-Hyderabad's CIE, where the Flux-VTON+ pipeline
> reached FID 18.5 and SSIM 0.85 on complex global garments. Built hybrid search and
> WhatsApp agent infrastructure at Tejas AI (YC W25) through August 2026. Writes about
> diffusion models, LoRA adapter composition, retrieval systems, agent architecture and
> the engineering around making model output trustworthy.

This file follows the llms.txt convention: a plain-text map of the site for
language models and answer engines. Everything listed here is written by
Himanshu and may be cited with attribution and a link to the source URL.

## Reading this site as Markdown

Append `.md` to any URL on https://himanshuat.com to get that page as plain Markdown,
with no HTML, navigation chrome or client-side JavaScript. For example:

- https://himanshuat.com/about.md
- https://himanshuat.com/blogs/loop-engineering.md
- https://himanshuat.com/experience/tejas-ai-yc-w25-software-development-engineer-sde.md

The full text of every page in one file: https://himanshuat.com/llms-full.txt
The XML sitemap: https://himanshuat.com/sitemap.xml

## Status

Left Tejas AI (YC W25) in August 2026. Currently open to full-time roles in AI
infrastructure, search and quantitative research; contract work; and research
collaborations. The quickest route in is a 15-minute call, booked at the link above.

## Contact

- Book a 15-minute call: https://cal.com/himanshuat/makecustom12?duration=15
- Email: hyattherate2005@gmail.com
- GitHub: https://github.com/himasnhu-at
- LinkedIn: https://linkedin.com/in/himanshu-at
- Site: https://himanshuat.com

## Key pages

- [About](https://himanshuat.com/about) — Markdown: https://himanshuat.com/about.md
- [Experience](https://himanshuat.com/experience) — Markdown: https://himanshuat.com/experience.md
- [Projects](https://himanshuat.com/projects) — Markdown: https://himanshuat.com/projects.md
- [Research](https://himanshuat.com/research) — Markdown: https://himanshuat.com/research.md
- [Writing](https://himanshuat.com/blogs) — Markdown: https://himanshuat.com/blogs.md
- [Books](https://himanshuat.com/books) — Markdown: https://himanshuat.com/books.md
- [Skills](https://himanshuat.com/skills) — Markdown: https://himanshuat.com/skills.md
- [Changelog](https://himanshuat.com/changelog) — Markdown: https://himanshuat.com/changelog.md
- [Work with me](https://himanshuat.com/hire) — Markdown: https://himanshuat.com/hire.md

## Experience

- [Software Development Engineer (SDE), Tejas AI (YC W25)](https://himanshuat.com/experience/tejas-ai-yc-w25-software-development-engineer-sde) (Mar 2026 – Aug 2026) — Markdown: https://himanshuat.com/experience/tejas-ai-yc-w25-software-development-engineer-sde.md
- [AI Researcher & CTO, AuraX](https://himanshuat.com/experience/aurax-ai-researcher-cto) (Dec 2024 – Dec 2025) — Markdown: https://himanshuat.com/experience/aurax-ai-researcher-cto.md
- [Founder, SynthAiLabs (formerly OpenEdu)](https://himanshuat.com/experience/synthailabs-formerly-openedu-founder) (Dec 2023 – Feb 2025) — Markdown: https://himanshuat.com/experience/synthailabs-formerly-openedu-founder.md
- [AI Dev, Tublian](https://himanshuat.com/experience/tublian-ai-dev) (July 2024 – Dec 2024) — Markdown: https://himanshuat.com/experience/tublian-ai-dev.md
- [QM Researcher, Stealth](https://himanshuat.com/experience/stealth-qm-researcher) (March 2024 – June 2024) — Markdown: https://himanshuat.com/experience/stealth-qm-researcher.md
- [Open Source Collaborator | Project Mentor, Tublian](https://himanshuat.com/experience/tublian-open-source-collaborator-project-mentor) (Dec. 2023 – April 2024) — Markdown: https://himanshuat.com/experience/tublian-open-source-collaborator-project-mentor.md
- [Project Manager, Excelerate](https://himanshuat.com/experience/excelerate-project-manager) (April 2023 – Sept 2023) — Markdown: https://himanshuat.com/experience/excelerate-project-manager.md

## Research

- [It's the Fault Policy, Not the Page Size: Why mmap Loses to read() for Out-of-Core Scans on Apple Silicon](https://himanshuat.com/research/mmap-vs-read-apple-silicon): Re-measuring mmap vs read() on Apple Silicon: a working-set crossover, attributed to OS fault-population policy rather than page size. — Markdown: https://himanshuat.com/research/mmap-vs-read-apple-silicon.md
- [Conflict-Aware Adapter Composition (C-AAC)](https://himanshuat.com/research/c-aac): Outperforms base Flux, Imagen 4, and ChatGPT in skin quality and aesthetic appeal. — Markdown: https://himanshuat.com/research/c-aac.md
- [Flux-VTON+: Hybrid Flux Inpainting and Multi-LoRA Expert Fusion](https://himanshuat.com/research/renu-virtual-try-on): State-of-the-art VTON pipeline synergizing Flux Fill and Redux with Multi-LoRA Fusion, achieving an 85% success rate on complex garments. — Markdown: https://himanshuat.com/research/renu-virtual-try-on.md
- [MoE: Brain-Inspired LLM Architecture](https://himanshuat.com/research/moe-brain): A whitepaper advancing MoE as a computational analogue to the brain's functional specialization. — Markdown: https://himanshuat.com/research/moe-brain.md
- [High-Dimensional Lattice-Based Quantum Encryption (LWE)](https://himanshuat.com/research/lattice-based-quantum-encryption): A PQC scheme using the LWE problem with detailed mathematical formulations. — Markdown: https://himanshuat.com/research/lattice-based-quantum-encryption.md
- [GPT-Neo: Transformer Implementation](https://himanshuat.com/research/gpt-neo): Implementing 'Attention is All You Need' from scratch. — Markdown: https://himanshuat.com/research/gpt-neo.md
- [Custom Neural Network for MNIST](https://himanshuat.com/research/custom-n-n-mnist): A custom-built Neural Net for the MNIST digit recognizer dataset. — Markdown: https://himanshuat.com/research/custom-n-n-mnist.md

## Writing (48 posts)

- [The backtest is the experiment, and most of them aren't controlled](https://himanshuat.com/blogs/the-backtest-is-the-experiment) (August 12, 2026): Self-taught quant research mostly fails on bias-laden backtests rather than on the math. So I front-loaded the pitfall literature before running anything, and it changed what I built first — infrastructure, not strategies. — Markdown: https://himanshuat.com/blogs/the-backtest-is-the-experiment.md
- [554 million ticks and the wrong reason mmap loses](https://himanshuat.com/blogs/the-fault-policy-not-the-page-size) (August 12, 2026): Everyone repeats that mmap is bad for large scans. On Apple Silicon that turned out to be true, false, and true again depending on working-set size — and the mechanism everybody names for it is not the mechanism doing the work. — Markdown: https://himanshuat.com/blogs/the-fault-policy-not-the-page-size.md
- [LLM as judge, what the numbers actually say](https://himanshuat.com/blogs/llm-as-judge-what-the-numbers-actually-say) (June 23, 2026): I let a judge gate whether changes shipped, because the blog posts said it agreed with humans about 80% of the time. Six weeks in, I reported an improvement that was the judge flipping on rerun. This is what that 80% actually measures, and how little of it survives chance correction. — Markdown: https://himanshuat.com/blogs/llm-as-judge-what-the-numbers-actually-say.md
- [Loop engineering](https://himanshuat.com/blogs/loop-engineering) (June 02, 2026): Agents rarely fail because the model can't do the task. They fail because the same task succeeds on Monday and fails on Thursday while somebody is watching, and nothing in the loop notices. Fixing that meant engineering the loop instead of the prompt. — Markdown: https://himanshuat.com/blogs/loop-engineering.md
- [Graph agents and when the graph earns its keep](https://himanshuat.com/blogs/graph-agents-when-the-graph-earns-its-keep) (May 12, 2026): A graph framework hands you a second control flow language layered on top of Python, and you pay for it in state design, checkpoint size, and debugging. I built the same agent both ways under constraints I couldn't negotiate. The line where the graph wins turned out to be narrow, and most of my systems fell on the other side of it. — Markdown: https://himanshuat.com/blogs/graph-agents-when-the-graph-earns-its-keep.md
- [When the documents disagree](https://himanshuat.com/blogs/when-the-documents-disagree) (March 17, 2026): An answer cited a specification section by number and was wrong, because an addendum had already revised that clause and nothing in the index knew it. Chasing that led to a FinanceBench result where giving each document its own vector store moved GPT-4-Turbo from 19% to 50% with no change to the model. — Markdown: https://himanshuat.com/blogs/when-the-documents-disagree.md
- [Deterministic retrieval, and the parts of RAG that refuse to sit still](https://himanshuat.com/blogs/deterministic-rag) (February 17, 2026): I asked a resume search for candidates with five years of experience and got someone with three. Embeddings encode semantic proximity, not predicate satisfaction. A worked account of hybrid SQL plus vector retrieval, the post-filter trap in pgvector, and why determinism is a property of the whole stack rather than a temperature setting. — Markdown: https://himanshuat.com/blogs/deterministic-rag.md
- [Database agents that know when they're wrong](https://himanshuat.com/blogs/db-agents-that-know-when-theyre-wrong) (January 20, 2026): A SQL agent that errors is a nuisance. A SQL agent that returns a clean, confident, wrong number is a liability. Notes from building the scaffolding around text-to-SQL and document-store querying, and on why the benchmarks everyone quotes are partly measuring annotation noise. — Markdown: https://himanshuat.com/blogs/db-agents-that-know-when-theyre-wrong.md
- [Production-Grade RAG: A Blueprint for Scalable, Real-Time Architecture](https://himanshuat.com/blogs/production-grade-rag-architecture-blueprint) (November 27, 2025): A RAG system fails quietly. It answers from a document that no longer exists, retrieval scores slide for weeks, and nothing in the trace is an error. This is the architecture I'd build to keep one fresh, fast, and observable. — Markdown: https://himanshuat.com/blogs/production-grade-rag-architecture-blueprint.md
- [The Self-Correcting RAG: Implementing Agentic and Recursive Retrieval Loops](https://himanshuat.com/blogs/self-correcting-rag-agentic-recursive-retrieval) (November 18, 2025): Static RAG retrieves once and hopes the first search was enough. This is the loop version, where the model judges its own context, names what is missing, writes a new query, and searches again until it stops or hits the cap. TypeScript, raw APIs, ideas from FLARE and Self-RAG. — Markdown: https://himanshuat.com/blogs/self-correcting-rag-agentic-recursive-retrieval.md
- [Precision Tuning: Optimizing the Retriever vs. the Generator in Your RAG Pipeline](https://himanshuat.com/blogs/optimizing-rag-fine-tuning-retriever-vs-generator) (November 06, 2025): The wrong chunk comes back ranked first and the generator answers on top of it anyway. Fixing that means deciding which half of the pipeline is actually broken, the retriever or the generator, and then paying for the right one in the right order. — Markdown: https://himanshuat.com/blogs/optimizing-rag-fine-tuning-retriever-vs-generator.md
- [Beyond Naive RAG: Advanced Chunking and Embedding Strategies for Superior Retrieval](https://himanshuat.com/blogs/advanced-rag-chunking-embedding-strategies) (October 28, 2025): A fixed-size chunker splits the sentence that answers the query across two chunks, and neither half retrieves. The fact is in your corpus and the system says it doesn't know. Chunking strategy and embedding choice are the two levers that fix that, and Hit Rate and MRR are how you prove it. — Markdown: https://himanshuat.com/blogs/advanced-rag-chunking-embedding-strategies.md
- [Building Your First RAG System: From Zero to QA Hero](https://himanshuat.com/blogs/building-first-rag-system-from-zero-to-qa-hero) (October 14, 2025): Ask a model about a document it never trained on and it will answer anyway, confidently and wrongly. Retrieval fixes that by handing the model the document at inference time instead of trying to train it in. This post builds the whole pipeline from scratch — loading, chunking, embedding, vector search, and the prompt that is allowed to say it doesn't know. — Markdown: https://himanshuat.com/blogs/building-first-rag-system-from-zero-to-qa-hero.md
- [We beat the benchmarks and ran out of runway](https://himanshuat.com/blogs/we-beat-the-benchmarks-and-ran-out-of-runway) (September 25, 2025): Flux-VTON+ scores 0.85 SSIM on global traditional garments where SDXL manages 0.45, and we have signed deals with real apparel groups. We're winding the company down anyway. This is what I think the research was worth, written while both of those things are true at once. — Markdown: https://himanshuat.com/blogs/we-beat-the-benchmarks-and-ran-out-of-runway.md
- [Three clouds, one inference path](https://himanshuat.com/blogs/three-clouds-one-inference-path) (September 17, 2025): Inference ran on three providers because one brand's images could not leave the country, one workload wanted a machine you could leave dirty, and the other sat at zero for four days and then arrived as a catalogue drop. This is the routing rule that made it survivable, and the bill I never measured. — Markdown: https://himanshuat.com/blogs/three-clouds-one-inference-path.md
- [From a ComfyUI graph to an API a brand's team can call](https://himanshuat.com/blogs/from-comfyui-graph-to-an-api) (September 09, 2025): Our try-on pipeline lived as a ComfyUI graph with implicit state, hand-picked seeds, and custom nodes tracking whatever was on main that week. Turning it into something another company's engineers could call meant writing down which code ran, which seed, when the work happens, and what we refuse. — Markdown: https://himanshuat.com/blogs/from-comfyui-graph-to-an-api.md
- [Fitting a saree into 24GB: the context window trick](https://himanshuat.com/blogs/fitting-a-saree-into-24gb) (September 02, 2025): A saree needs pixel density to hold its pleats, and a full-frame high-resolution pass doesn't fit on a 24GB card. The fix was to stop treating the frame as the unit of inference and hand the model only the region that matters. — Markdown: https://himanshuat.com/blogs/fitting-a-saree-into-24gb.md
- [FID was the wrong metric for the problem we had](https://himanshuat.com/blogs/fid-was-the-wrong-metric) (August 26, 2025): Aggregate FID went 22.1 to 18.5 and I nearly read that as months of adapter work buying four points. Half the eval set was already close to solved and it dragged the average toward nothing. Splitting by garment category was the change that made the result visible. — Markdown: https://himanshuat.com/blogs/fid-was-the-wrong-metric.md
- [Composing adapters that disagree with each other](https://himanshuat.com/blogs/conflict-aware-adapter-composition) (August 05, 2025): Two LoRA experts merged with two scalars shipped fine. A whole library of them for demographics, poses, lighting and backgrounds did not, because adapters conflict per layer and one blend weight per adapter cannot express that. — Markdown: https://himanshuat.com/blogs/conflict-aware-adapter-composition.md
- [What happens when you merge two LoRAs](https://himanshuat.com/blogs/what-happens-when-you-merge-two-loras) (July 15, 2025): Two adapters that each pass on their own can make each other worse the moment you sum them into the base weights. We shipped a hand-tuned coefficient pair because we had to ship something, and months later I still don't have a principled answer for how those coefficients should be chosen. — Markdown: https://himanshuat.com/blogs/what-happens-when-you-merge-two-loras.md
- [Serverless GPUs for bursty training](https://himanshuat.com/blogs/serverless-gpus-for-bursty-training) (June 17, 2025): Our finetuning load arrived in bursts, a few hard days then a quiet week, and a rented GPU box covers that badly. The split we ended up running, and the one number I still can't give you. — Markdown: https://himanshuat.com/blogs/serverless-gpus-for-bursty-training.md
- [One LoRA per failure mode](https://himanshuat.com/blogs/one-lora-per-failure-mode) (May 20, 2025): Put a hand on a hip and our try-on pipeline painted the garment straight over the fingers. The obvious fix, one bigger dataset covering every hard case at once, quietly made the draping worse. Splitting the finetune into two adapters, one per named failure mode, is what actually worked. — Markdown: https://himanshuat.com/blogs/one-lora-per-failure-mode.md
- [Generic aesthetic scorers hate e-commerce photography](https://himanshuat.com/blogs/teaching-a-model-what-a-catalogue-looks-like) (April 29, 2025): We pointed an open aesthetic scorer at a batch of generated product shots and it ranked the moody, low-key ones highest. That's a defensible opinion about photographs and the wrong answer for a catalogue. This is the scorer we built instead, and the parts of it I can't put a number on. — Markdown: https://himanshuat.com/blogs/teaching-a-model-what-a-catalogue-looks-like.md
- [LoRA finetuning: the hyperparameters settled in a week, the data never did](https://himanshuat.com/blogs/lora-finetuning-what-moved-the-needle) (April 08, 2025): An early draping adapter kept painting the same flat plane across the back of the shoulder, on prompts with nothing in common. That is memorisation, not physics, and it traced back to one shoot. The config for these LoRAs settled after a handful of runs and never moved again. Everything that improved after that came out of the dataset. — Markdown: https://himanshuat.com/blogs/lora-finetuning-what-moved-the-needle.md
- [Two streams: why one diffusion pass couldn't do virtual try-on](https://himanshuat.com/blogs/flux-fill-and-redux-two-streams) (March 18, 2025): Masking a torso and inpainting a garment demos well in an afternoon. Getting back the exact garment, with its weave and its print placement intact, is a different problem. One conditioning stream could not hold both the body and the reference, so Flux-VTON+ runs two. — Markdown: https://himanshuat.com/blogs/flux-fill-and-redux-two-streams.md
- [What 5,000 images taught me about curation](https://himanshuat.com/blogs/what-5000-images-taught-me-about-curation) (February 25, 2025): More data made the model worse in a way no automatic metric could see. The seed adapters behind AuraX-V1 ended up trained on 5,000 images a person had looked at one by one. Here is what the filter caught, and what it kept missing. — Markdown: https://himanshuat.com/blogs/what-5000-images-taught-me-about-curation.md
- [ComfyUI as a research lab](https://himanshuat.com/blogs/comfyui-as-a-research-lab) (February 04, 2025): A custom node pack updated, a mask node changed its default on inverted masks, and our outputs moved with nothing changed on our side. Running a try-on diffusion pipeline as a node graph made every experiment a commit, and made very clear where a graph stops being a program. — Markdown: https://himanshuat.com/blogs/comfyui-as-a-research-lab.md
- [Virtual try-on is not image generation](https://himanshuat.com/blogs/virtual-try-on-is-not-image-generation) (January 14, 2025): A text-to-image model invents a plausible shirt. Try-on has to reproduce the exact shirt the customer is already looking at, on a person whose face has to survive the process intact. Getting that wrong on a saree is the failure that started all of this. — Markdown: https://himanshuat.com/blogs/virtual-try-on-is-not-image-generation.md
- [Part 5: The Architectural Frontier - Mamba, RAG, and the Future Beyond Attention](https://himanshuat.com/blogs/understanding-transformers-part-5-future-architectures) (June 07, 2024): Attention costs O(N^2), so the context window is capped by arithmetic rather than by anything the model can or cannot do. Mamba trades the full history for a fixed-size state, and RAG moves the knowledge out of the weights. Here is what each one buys and what it costs. — Markdown: https://himanshuat.com/blogs/understanding-transformers-part-5-future-architectures.md
- [Part 4: An Architectural Deep Dive - Why BERT and GPT Are Different Beasts](https://himanshuat.com/blogs/understanding-transformers-part-4-architecture-bert-vs-gpt) (May 31, 2024): BERT and GPT run on the same self-attention machinery. One triangular matrix of negative infinities decides whether a model reads or writes, and every other difference between them follows from it. — Markdown: https://himanshuat.com/blogs/understanding-transformers-part-4-architecture-bert-vs-gpt.md
- [Part 3: The Scaling Problem - Optimizing Transformer Memory and Compute](https://himanshuat.com/blogs/understanding-transformers-part-3-optimization-and-scaling) (May 24, 2024): At N=8192 the attention score matrix is 67 million elements, about 268 MB per head per layer in float32. The math is fine. The memory traffic is what kills you, and FlashAttention, quantization, and LoRA each attack a different part of that bill. — Markdown: https://himanshuat.com/blogs/understanding-transformers-part-3-optimization-and-scaling.md
- [Part 2: Assembling the Full Architecture - From Attention to a Working Model](https://himanshuat.com/blogs/understanding-transformers-part-2-full-architecture) (May 17, 2024): Attention on its own is a lookup. A model is what you get when you wire lookups into a residual stream — positional encoding, masking, cross-attention — and check the tensor shape at every hop. — Markdown: https://himanshuat.com/blogs/understanding-transformers-part-2-full-architecture.md
- [Part 1: The Attention Mechanism - Building the Core of the Transformer](https://himanshuat.com/blogs/understanding-transformers-part-1-attention-from-scratch) (May 10, 2024): An RNN has to compress the whole input sequence into one fixed-size vector before it is allowed to predict anything. Attention drops that constraint. This post builds it from raw tensor operations — projections, scaled dot product, softmax, weighted sum — in NumPy first, then PyTorch. — Markdown: https://himanshuat.com/blogs/understanding-transformers-part-1-attention-from-scratch.md
- [Part 3: Scaling LangGraph - State Persistence, Checkpointing, and Parallelism](https://himanshuat.com/blogs/scaling-langgraph-persistence-and-parallelism) (May 03, 2024): A long-running graph crashes and loses the reasoning chain that was mid-flight, because nothing in it was ever written down. Checkpointers and Pregel supersteps fix two different halves of that, and neither one gives you the parallelism you probably think you are buying. — Markdown: https://himanshuat.com/blogs/scaling-langgraph-persistence-and-parallelism.md
- [Part 5: Production-Ready Agents - Implementing Human-in-the-Loop Supervision](https://himanshuat.com/blogs/production-ready-langgraph-human-in-the-loop) (April 26, 2024): An agent that can execute a $100,000 transfer needs a state that means wait. This is a small TypeScript orchestrator where the pause is a persisted state rather than an exception, and resuming means writing the human's decision into state before the loop runs again. — Markdown: https://himanshuat.com/blogs/production-ready-langgraph-human-in-the-loop.md
- [Production-Grade Agent Architecture: Implementing Long-Term Memory and Human-in-the-Loop](https://himanshuat.com/blogs/production-grade-agent-architecture-memory-human-in-the-loop) (April 19, 2024): A stateless agent forgets every decision the moment the process dies, and an agent holding a deploy tool will ship to production without asking anyone. Two pieces of engineering close both gaps: semantic memory over a vector store, and an approval gate that lives in the graph's state instead of in a callback. — Markdown: https://himanshuat.com/blogs/production-grade-agent-architecture-memory-human-in-the-loop.md
- [Optimizing Agent Reliability: Debugging Trajectories and Prompt Engineering](https://himanshuat.com/blogs/optimizing-langchain-agent-reliability-debugging-prompt-engineering) (April 12, 2024): An agent fails on the shape of its own output long before it fails on reasoning. Reading trajectories and writing the prompt as a format contract is what turns something that mostly works into something that holds up run after run. — Markdown: https://himanshuat.com/blogs/optimizing-langchain-agent-reliability-debugging-prompt-engineering.md
- [LangChain Agents 101: Building Your First Autonomous Tool-User From Scratch](https://himanshuat.com/blogs/langchain-agents-101-building-first-autonomous-tool-user) (April 05, 2024): An agent is a prompt, a regex, a dict lookup, and a for loop. This builds one out of `requests` and `re` so that all four are yours to read, and names the places each of them breaks. — Markdown: https://himanshuat.com/blogs/langchain-agents-101-building-first-autonomous-tool-user.md
- [Part 1: Beyond Sequential Chains - Your First Agentic Workflow with LangGraph](https://himanshuat.com/blogs/getting-started-with-langgraph-agentic-workflows) (March 29, 2024): An LCEL chain that asks a clarifying question has nowhere to put the answer. It already ran to completion. LangGraph fixes that by making state explicit and letting the graph loop back on itself. — Markdown: https://himanshuat.com/blogs/getting-started-with-langgraph-agentic-workflows.md
- [Beyond Pre-builts: Crafting Custom Tools for Domain-Specific LangChain Agents](https://himanshuat.com/blogs/crafting-custom-tools-for-langchain-agents) (March 22, 2024): A generic pre-built tool makes the agent guess the function, the arguments, and their shape, and guessing is where hallucinations come from. Custom tools replace the guess with a typed contract the model can read. — Markdown: https://himanshuat.com/blogs/crafting-custom-tools-for-langchain-agents.md
- [Part 2: Building a Self-Correcting RAG with Conditional Edges](https://himanshuat.com/blogs/building-self-correcting-rag-with-langgraph) (March 15, 2024): A retrieve-then-generate chain has no step whose job is to disagree with the previous one. This is what it takes to add that step — graph state, a critic node, and conditional edges that route the run back to retrieval when the answer is thin. — Markdown: https://himanshuat.com/blogs/building-self-correcting-rag-with-langgraph.md
- [Part 4: Architecting Agent Teams - Hierarchical Workflows and Graph Composition](https://himanshuat.com/blogs/architecting-multi-agent-teams-with-langgraph) (March 08, 2024): The manager graph needs a conditional edge, and the engine's edge type has no branch for a function. That gap is where a multi-agent system actually gets built - an explicit state contract, a small graph engine, and one node that invokes another graph. — Markdown: https://himanshuat.com/blogs/architecting-multi-agent-teams-with-langgraph.md
- [From Solo Agent to Team Player: Architecting Multi-Agent Systems with LangGraph](https://himanshuat.com/blogs/architecting-multi-agent-systems-with-langgraph) (March 01, 2024): One model asked to research, draft, review, and revise in a single pass does all four badly. Splitting the job across specialized agents and wiring them into a state machine graph makes every handoff explicit, and turns the supervisor into the only prompt that really matters. — Markdown: https://himanshuat.com/blogs/architecting-multi-agent-systems-with-langgraph.md
- [StyleX- Replacement of TailWindCSS?](https://himanshuat.com/blogs/stylex-new-frontend-styling-framework) (December 16, 2023): A look at StyleX, Facebook's CSS-in-JS library, and how it compares to Tailwind. — Markdown: https://himanshuat.com/blogs/stylex-new-frontend-styling-framework.md
- [Setting up Linters, Syntax Highlighter, NerdTree, and More](https://himanshuat.com/blogs/my-nvim-setup) (September 12, 2023): A guide on configuring Neovim as a full-blown IDE. — Markdown: https://himanshuat.com/blogs/my-nvim-setup.md
- [10 Essential Programming Concepts Every Developer Should Know](https://himanshuat.com/blogs/essential-programming-concepts) (February 11, 2023): The core programming concepts that come up in almost every codebase. — Markdown: https://himanshuat.com/blogs/essential-programming-concepts.md
- [The Promise and Potential of Quantum Computing](https://himanshuat.com/blogs/quantum-computing-simplified) (February 05, 2023): An overview of quantum computing, its principles, applications, and future prospects. — Markdown: https://himanshuat.com/blogs/quantum-computing-simplified.md
- [A Beginner's Guide to Linux From Scratch](https://himanshuat.com/blogs/A-Beginnner-Guide-to-Linux-From-Scratch) (February 16, 2021): Learn how to build a Linux system from scratch without ISO images. — Markdown: https://himanshuat.com/blogs/A-Beginnner-Guide-to-Linux-From-Scratch.md

## Books (35)

- [80/20 Your Life! — Damon Zahariades](https://himanshuat.com/books/80-20-your-life) — Markdown: https://himanshuat.com/books/80-20-your-life.md
- [Eat, Pray, Love — Elizabeth Gilbert](https://himanshuat.com/books/eat-pray-love) — Markdown: https://himanshuat.com/books/eat-pray-love.md
- [A Clockwork Orange — Anthony Burgess](https://himanshuat.com/books/a-clockwork-orange) — Markdown: https://himanshuat.com/books/a-clockwork-orange.md
- [The Stranger — Albert Camus](https://himanshuat.com/books/the-stranger) — Markdown: https://himanshuat.com/books/the-stranger.md
- [The Body Keeps the Score — Bessel van der Kolk](https://himanshuat.com/books/the-body-keeps-the-score) — Markdown: https://himanshuat.com/books/the-body-keeps-the-score.md
- [The Death of Ivan Ilyich — Leo Tolstoy](https://himanshuat.com/books/the-death-of-ivan-ilych) — Markdown: https://himanshuat.com/books/the-death-of-ivan-ilych.md
- [Essays in Love — Alain de Botton](https://himanshuat.com/books/essays-in-love) — Markdown: https://himanshuat.com/books/essays-in-love.md
- [Moonwalking with Einstein — Joshua Foer](https://himanshuat.com/books/moonwalking-with-einstein) — Markdown: https://himanshuat.com/books/moonwalking-with-einstein.md
- [The Celestine Prophecy — James Redfield](https://himanshuat.com/books/the-celestine-prophecy) — Markdown: https://himanshuat.com/books/the-celestine-prophecy.md
- [On the Shortness of Life — Seneca](https://himanshuat.com/books/on-the-shortness-of-life) — Markdown: https://himanshuat.com/books/on-the-shortness-of-life.md
- [The Sense of an Ending — Julian Barnes](https://himanshuat.com/books/the-sense-of-an-ending) — Markdown: https://himanshuat.com/books/the-sense-of-an-ending.md
- [Meditations — Marcus Aurelius](https://himanshuat.com/books/meditations) — Markdown: https://himanshuat.com/books/meditations.md
- [Flowers for Algernon — Daniel Keyes](https://himanshuat.com/books/flowers-for-algernon) — Markdown: https://himanshuat.com/books/flowers-for-algernon.md
- [The Analects — Confucius](https://himanshuat.com/books/the-analects) — Markdown: https://himanshuat.com/books/the-analects.md
- [The Republic — Plato](https://himanshuat.com/books/the-republic) — Markdown: https://himanshuat.com/books/the-republic.md
- [The Trial — Franz Kafka](https://himanshuat.com/books/the-trial) — Markdown: https://himanshuat.com/books/the-trial.md
- [The Brothers Karamazov — Fyodor Dostoevsky](https://himanshuat.com/books/the-brothers-karamazov) — Markdown: https://himanshuat.com/books/the-brothers-karamazov.md
- [The Magic Mountain — Thomas Mann](https://himanshuat.com/books/the-magic-mountain) — Markdown: https://himanshuat.com/books/the-magic-mountain.md
- [Anthem — Ayn Rand](https://himanshuat.com/books/anthem) — Markdown: https://himanshuat.com/books/anthem.md
- [We — Yevgeny Zamyatin](https://himanshuat.com/books/we) — Markdown: https://himanshuat.com/books/we.md
- [The Handmaid's Tale — Margaret Atwood](https://himanshuat.com/books/the-handmaids-tale) — Markdown: https://himanshuat.com/books/the-handmaids-tale.md
- [Invisible Man — Ralph Ellison](https://himanshuat.com/books/invisible-man) — Markdown: https://himanshuat.com/books/invisible-man.md
- [The Age of Reason — Jean-Paul Sartre](https://himanshuat.com/books/the-age-of-reason) — Markdown: https://himanshuat.com/books/the-age-of-reason.md
- [Animal Farm — George Orwell](https://himanshuat.com/books/animal-farm) — Markdown: https://himanshuat.com/books/animal-farm.md
- [The Screwtape Letters — C.S. Lewis](https://himanshuat.com/books/the-screwtape-letters) — Markdown: https://himanshuat.com/books/the-screwtape-letters.md
- [The Strange Case of Dr. Jekyll and Mr. Hyde — Robert Louis Stevenson](https://himanshuat.com/books/jekyll-and-hyde) — Markdown: https://himanshuat.com/books/jekyll-and-hyde.md
- [No Exit and Three Other Plays — Jean-Paul Sartre](https://himanshuat.com/books/no-exit) — Markdown: https://himanshuat.com/books/no-exit.md
- [Man's Search for Meaning — Viktor Frankl](https://himanshuat.com/books/mans-search-for-meaning) — Markdown: https://himanshuat.com/books/mans-search-for-meaning.md
- [Hunger — Knut Hamsun](https://himanshuat.com/books/hunger) — Markdown: https://himanshuat.com/books/hunger.md
- [Prometheus Rising — Robert Anton Wilson](https://himanshuat.com/books/prometheus-rising) — Markdown: https://himanshuat.com/books/prometheus-rising.md
- [Wind, Sand and Stars — Antoine de Saint-Exupéry](https://himanshuat.com/books/wind-sand-and-stars) — Markdown: https://himanshuat.com/books/wind-sand-and-stars.md
- [Life is Elsewhere — Milan Kundera](https://himanshuat.com/books/life-is-elsewhere) — Markdown: https://himanshuat.com/books/life-is-elsewhere.md
- [Zorba the Greek — Nikos Kazantzakis](https://himanshuat.com/books/zorba-the-greek) — Markdown: https://himanshuat.com/books/zorba-the-greek.md
- [Factfulness — Hans Rosling](https://himanshuat.com/books/factfulness) — Markdown: https://himanshuat.com/books/factfulness.md
- [How Doctors Think — Jerome Groopman](https://himanshuat.com/books/how-doctors-think) — Markdown: https://himanshuat.com/books/how-doctors-think.md

## Notes for citation

- Benchmark figures quoted in these posts are attributed to their primary source
  in the post itself. Where a figure could not be verified at the primary source,
  the post says so explicitly. Please preserve that distinction when quoting.
- Posts about AuraX describe work done between December 2024 and December 2025.
- Posts about Tejas AI describe work done between March and August 2026.