Anand is available for hire
Hire AnandAnand Ramanathan
Verified Expert in Engineering
Machine Learning Developer
Bellevue, WA, United States
Toptal member since April 29, 2020
Bio
Anand (RC) is a leading full-stack AI engineer, building AI-based experiences with the highest-level AI coding assistance. He combines 20+ years of engineering (Microsoft, Amazon, startups) with hands-on expertise in all things AI. His volume (breadth x depth) in all things AI is in the top 1% worldwide. He is the go-to person for AI advice in all his circles. He builds and releases high-quality apps with AI.
Portfolio
MLAI, LLC.
Natural Language Processing (NLP), SaaS, Machine Learning, RESTful Development...
Ripcord
Generative Pre-trained Transformers (GPT), Chatbots...
Healthcare Provider Client (via Toptal)
Machine Learning, Deep Learning, Computer Vision, Healthcare...
Experience
- Artificial Intelligence (AI) - 10 years
- AI Research - 9 years
- Vector Stores - 4 years
- Prompt Engineering - 3 years
- AI Agents - 3 years
- Ollama - 3 years
- OpenAI Assistants API - 1 year
- ChatGPT - 1 year
Preferred Environment
Python, Ollama, Claude Code, Pi.dev, Pydantic, Agentic AI, OpenRouter, Retrieval-augmented Generation (RAG), MacOS, Obsidian, Machine Learning, Artificial Intelligence (AI), Model Context Protocol (MCP), FastAPI, Data Engineering, Optical Character Recognition (OCR), SQL, API Design, Error Handling, Testing, API Integration, LLM Reasoning, Vector Databases, Anthropic, Conversational AI, Small Language Models (SLMs)
The most amazing...
...thing I've created is Maibook - a private, proactive, and reactive personalized multi-agent AI community based on user activity: https://maibook.app.
Work Experience
Founder | AI Engineer
2020 - PRESENT
MLAI, LLC.
- Built Maibook, a multi-agent, local-first private AI community: a cross-platform desktop app that dynamically creates 100s of personalized agents from the user's activity and augments their interests and content.
- Built OllamaDash, a live LLM intelligence dashboard bwith 20,000+ models from Hugging Face, Ollama, OpenRouter, and Artificial Analysis, with cross-source search, filtering, and a radar chart comparing 15 benchmarks.
- Designed a statistically grounded price-performance metric (per-benchmark OLS fit of score vs log-price, min-max normalized) that auto-recalibrates as the market moves; wrote the formal spec and mirrored it in Python for live-data tests.
- Developed a fully automated agentic AI pipeline that generates educational comics from a Wikipedia page; created 300+ comics, published 10 on the Kindle store and several on Gumroad.
- Built Maiweb, a live, highly customizable feed into news, blogs, papers, and videos on the web.
- Built ahai (https://ahai.app), a native Mac app that uses Apple MLX to scan markdown files and surface forward-looking ideas (features to build, content to create, ideas to explore); 100% local and privacy-first.
- Built an opinionated Claude Code CLI visualizer: a single-file HTML app that turns local Claude Code history into a browsable interface with thinking blocks, tool-call visualization, diff views, search, and Markdown export (available on Gumroad).
- Built the AI Apps Explorer, a single-page catalog of LLM and agentic apps across categories. Try it: https://gisthost.github.io/?7023b76dde12cbb8f37deccfb122d49d/ai-apps-explorer.html.
- Built a standalone local viewer for Allen AI Asta's deep-research reports. Try it: https://gisthost.github.io/?e647e0142cff6e817deafe5eed73b2fb/asta_viewer.html.
- Extended nanocode, a minimal single-file Claude Code alternative, with LiteLLM support for local models via Ollama and LM Studio, plus keyboard navigation: https://github.com/rcanand/nanocode_litellm.
Technologies: Natural Language Processing (NLP), SaaS, Machine Learning, RESTful Development, Python, Data Science, Benchmarking, Ollama, Open WebUI, Agentic AI, OpenRouter, Agentic Coding, Agentic Frameworks, Agentic AI Systems, AI Agents, Agentic Workflow Design, PySide, RAG Systems, Artificial Intelligence (AI), Large Language Models (LLMs), Open-source LLMs, Prompt Engineering, Context Engineering, Loop Engineering, Harness Engineering, Foundation Models, Claude Code, Claude, Pi Dev, FastAPI, FastMCP, Hugging Face, Model Evaluation, Metrics, AutoML, Generative Artificial Intelligence (GenAI), Deep Learning, OpenAI, OpenAPI, LiteLLM, Nanocode, Meta Llama, Qwen, MLX, Quantization, Stable Diffusion, Z Image Turbo, Blackforest Labs Flux, Automatic1111, Pydantic, uv, Tauri, JavaScript, Flask, APIs, REST, RESTFul APIs, REST APIs, Web Scraping, RSS Feeds, Vector Stores, Vector Search, Embedding Models, Evaluation, Datasets, Data Analysis, Research, AI Research, AI Tools, Pi.dev, Obsidian, Obsidian Plugins, Browser Plugins, Software Architecture, Architecture, Full-stack, Full-stack Development, MacOS, Chatbots, AI Chatbots, Model Context Protocol (MCP), LangChain, Data Engineering, Optical Character Recognition (OCR), SQL, API Design, Error Handling, Remote Work, Testing, Software Engineering
Principal Machine Learning Scientist
2022 - 2024
Ripcord
- Built and launched Docufai, a v1 web application to chat with documents using generative AI. I owned all aspects of the AI—from experiments and benchmarking to production AI code. Also influenced product, engineering, design, strategy, and release.
- Added vector stores/retrieval-augmented generation (RAG) as a key component. Created an in-memory RAG index. Researched and provided several alternatives for integrating a vector store and search into existing text search databases and ecosystems.
- Retrained a neural network with new data to split, classify, and extract key value pairs from documents—from dataset creation and curation to training.
- Reviewed and provided guidance on the training to launch the pipeline of an object detection model (YOLO-based).
- Contributed to a human-in-the-loop AI model that provided a model-based first cut of annotations (key-value pairs) in documents that were then reviewed and updated by humans.
- Researched several ways to make LLM/GPT-based apps work correctly—hallucination reduction, suggesting questions in documents, checking AI-generated answers using AI, and more. Many of these were later validated by public external research.
- Evaluated and implemented various strategies to deal with large and numerous documents when interacting with LLMs like ChatGPT.
- Investigated custom GPTs and the Assistants API from OpenAI to identify how to use these in our products.
- Researched and shortlisted transformer-based models (LayoutLM, LILT, etc.) for document layout models that incorporate computer vision and language, NLP, and NLU for handling scanned documents and images containing text.
- Evaluated and used several OCR libraries for extracting text from documents.
Technologies: Generative Pre-trained Transformers (GPT), Chatbots, Chatbot Conversation Design, Large Language Models (LLMs), Retrieval-augmented Generation (RAG), OpenAI, OpenAI GPT-4 API, OpenAI GPT-3 API, PaLM 2, Google, Gecko, Embedding Models, Claude, LangChain, ChromaDB, Qdrant, AI Research, Research, Benchmarking, Datasets, Evaluation, Azure Machine Learning, Azure Cognitive Services, Artificial Intelligence (AI), Python 3, Python API, Vector Search, Hybrid Search, Prompt Engineering, AI Agents, Amazon Bedrock AgentCore, Amazon SageMaker, Llama 2, MTEB, Large Model Systems Organization (LMSYS), OpenAI Assistants API, Google Colaboratory (Colab), Machine Learning, Deep Learning, Natural Language Processing (NLP), Full-stack, PyTorch, NumPy, Pandas, Exploratory Data Analysis, Data Science, Seaborn, Jupyter, Computer Vision, Azure IaaS, TensorFlow, Python, Google Cloud Platform (GCP), Deep Neural Networks (DNNs), Algorithms, MacOS, Visual Studio Code (VS Code), PostgreSQL, APIs, Architecture, Convolutional Neural Networks (CNNs), Neural Networks, Recurrent Neural Networks (RNNs), Sequence Models, Hyperparameters, Regularization, Natural Language Toolkit (NLTK), SpaCy, Named-entity Recognition (NER), Mathematics, Language Models, ChatGPT, Vector Stores, Full-stack Development, iPaaS, Generative Artificial Intelligence (GenAI), Generative Pre-trained Transformer 3 (GPT-3), Data Analysis, FastAPI, Data Engineering, Optical Character Recognition (OCR), SQL, Agentic Frameworks, API Design, Error Handling, Remote Work, Testing, API Integration, Communication, Data Pipelines, Fine-tuning, Vector Databases, Documentation, RAG Architecture, Anthropic, Conversational AI, Multimodal GenAI, Technical Architecture, Software Engineering
AI/ML Engineer
2022 - 2022
Healthcare Provider Client (via Toptal)
- Conducted feasibility analysis of an ML model for the business problem.
- Created estimates of data collection and annotation needed.
- Provided candidate models to evaluate once data was available.
- Answered all client's questions to their total satisfaction.
Technologies: Machine Learning, Deep Learning, Computer Vision, Healthcare, 3D Image Processing, Image Processing, Medical Imaging, Task Analysis, Google Colaboratory (Colab), Research, Datasets, Python API, Data Analysis, Artificial Intelligence (AI), Data Engineering, SQL, API Design, Error Handling, Remote Work, Testing, Neural Networks, API Integration, Communication, Data Pipelines, Documentation, Technical Architecture, Software Engineering
Senior AI Engineer
2021 - 2022
RedRoute
- Owned AI for the company; defined the AI roadmap for the company, combining several ideas from business, product, and technology as well as from academia, research, and industry.
- Led the audio-related data science roadmap and work; mentored/technically managed an experienced audio data science researcher.
- Built a real time audio barge-in detection model that performed very well with low resource requirements, compared to several prior attempts.
- Improved the performance of an eCommerce intent detection model by 2-5%—analyzing and relabeling the data, manually prioritizing and labeling high impact utterances, retraining and redeploying the model, and creating and monitoring looker dashboards.
- Increased customer handle rates by adding richer responses for eCommerce customers; this was done by answering frequently asked questions based on information from the customer's website and knowledge base. Created dashboards to monitor these changes.
- Created a transcriber/speaker diary tool to transcribe conversations.
- Built several dashboards and looks (queries) in Looker to monitor the impact of different changes.
- Educated the team on Kanban and influenced the deployment of a variant of Kanban in the company.
- Contributed to other areas of the company, including preparing to scale and establishing and improving processes and the business/sales/marketing/product strategy.
Technologies: Python, Python 3, TensorFlow, Deep Learning, Real-time Data, Interactive Voice Response (IVR), Dialog Systems, Amazon Web Services (AWS), Ansible, Docker, Machine Learning Operations (MLOps), Google Speech-to-Text, Labeling, Jupyter Notebook, Jupyter, Kanban, Looker, ETL, MongoDB, Snowflake, Flask, Technical Hiring, Source Code Review, Code Review, Task Analysis, Interviewing, Team Management, Hyperparameters, Regularization, Language Models, Google Colaboratory (Colab), Chatbot Conversation Design, Chatbots, Large Language Models (LLMs), Research, Benchmarking, Datasets, Python API, Full-stack Development, Conversational Agent, Data Analysis, Machine Learning, Artificial Intelligence (AI), Data Engineering, SQL, API Design, Error Handling, Remote Work, Testing, Neural Networks, API Integration, Communication, Data Pipelines, Documentation, Conversational AI, Multimodal GenAI, Technical Architecture, Text-to-Speech (TTS), Software Engineering
Data Scientist
2018 - 2020
Microsoft
- Built models to evaluate over 100 datasets to discover methods to improve the AutoML library.
- Improved the product on the benchmark against the competition by investigating and explaining a competing AutoML product's scoring methodology for imbalanced multi-class classification. I did this by digging deep into how metrics were computed.
- Analyzed feature importances by training over 100 datasets with automatic feature engineering libraries to prioritize featurizers for our AutoML product.
- Developed an end-to-end AutoML framework as part of a Hackathon project to understand what approaches would work best for AutoML.
- Delivered a dataset analysis and onboarding tool, which enabled evaluating, filtering, cleaning, and onboarding of 20 datasets into our benchmarking corpus.
- Built better complete performance graphs to improve confidence in the performance of each competitor across a corpus of over 100 datasets.
- Identified gaps in our benchmarking dataset corpus distribution and added over 20 datasets to fill those gaps.
Technologies: R, Microsoft Power BI, Matplotlib, Scikit-learn, AutoML, Azure, Jupyter, Pandas, NumPy, Python, JSON REST APIs, REST APIs, SaaS, Machine Learning, Exploratory Data Analysis, Python 3, RESTful Development, Docker, SQL, Statistical Learning, Deep Learning, Data Science, Visual Studio Code (VS Code), Artificial Intelligence (AI), Seaborn, Azure IaaS, TensorFlow, Keras, Algorithms, Linux, Windows, APIs, Architecture, Artificial Neural Networks (ANN), SciPy, Tidyverse, Ggplot2, Systems, Neural Networks, Azure Machine Learning, Flask, Agile, Source Code Review, Code Review, Task Analysis, Hyperparameters, Regularization, Research, Benchmarking, Datasets, Evaluation, Python API, Data Analysis, Data Engineering, API Design, Error Handling, Testing, API Integration, Communication, Data Pipelines, Fine-tuning, Documentation, Technical Architecture
Founder
2015 - 2019
Meon
- Built a web platform that allows you to create apps in minutes, Meonapp.com.
- Developed 80 apps in two days across several application domains.
- Created apps for various clients on the platform, including a tailor management app, a music composition app.
- Added capabilities for both general users and developers to build apps.
- Enabled apps to be hosted as soon as built, reducing turnaround time greatly.
Technologies: Heroku, PostgreSQL, Ruby on Rails (RoR), Ruby, SaaS, RESTful Development, jQuery, JavaScript, SQL, REST, Full-stack, Visual Studio Code (VS Code), Algorithms, Linux, MacOS, Java, APIs, Architecture, Systems, REST APIs, Source Code Review, Code Review, Task Analysis, Research, Full-stack Development, iPaaS, API Design, Error Handling, Testing, API Integration, Data Pipelines, Documentation, Technical Architecture, Software Engineering
Senior Software Engineer
2017 - 2018
Divensi, Inc.
- Researched and developed a deep learning model for 3D point cloud semantic segmentation of imbalanced outdoor Lidar data, with near state-of-the-art results for outdoor Lidar.
- Developed a V1 cloud-hosted decision support system web application utilized by several enterprise users for a remote startup client. Hired a technical team and transitioned the product. Offered a CTO position by the client CEO.
- Built a data pipeline framework for machine learning experimentation with Lidar data.
Technologies: PDAL, Laspy, LiDAR, Jupyter, TensorFlow, Python, JSON REST APIs, Machine Learning, Exploratory Data Analysis, Python 3, Pandas, RESTful Development, JavaScript, Google Cloud Platform (GCP), SQL, Deep Learning, Statistical Learning, NumPy, Data Science, REST, Fast.ai, Full-stack, PostgreSQL, Artificial Intelligence (AI), Computer Vision, Keras, Django, Django REST Framework, Deep Neural Networks (DNNs), Algorithms, MacOS, APIs, Architecture, Artificial Neural Networks (ANN), Image Recognition, Matplotlib, Scikit-learn, Convolutional Neural Networks (CNNs), SciPy, Systems, Neural Networks, Agile, REST APIs, 3D Image Processing, Image Processing, Technical Hiring, Source Code Review, Code Review, Task Analysis, Interviewing, Team Management, Mathematics, Research, Datasets, Evaluation, Python API, Full-stack Development, iPaaS, API Design, Error Handling, Remote Work, Testing, API Integration, Communication, Data Pipelines, Fine-tuning, Documentation, Technical Architecture, Software Engineering
Freelance Developer
2012 - 2014
Self Employment
- Built nine educational games that were released to the App Store; iOS and cross-platform using the Corona SDK.
- Performed App Store optimization to maximize adoption and saw over 30,000 downloads across games.
- Developed a broad range of games, from running quizzes to new mathematical puzzles. The samurai game was highly appreciated by middle school teachers in the US.
Technologies: ASP.NET, Corona SDK, Objective-C, REST APIs, SQL, iOS, Algorithms, MacOS, Architecture, Systems, Task Analysis, Full-stack Development, Machine Learning, Artificial Intelligence (AI), API Design, Error Handling, Remote Work, Testing, API Integration, Technical Architecture, Software Engineering
Founder
2009 - 2011
Thouwords, LLC.
- Built a web application to make a textual website more visual by performing topic modeling with Alchemy API. Obtained images for each topic from Wikipedia APIs.
- Created a Wikipedia visual navigator using topic modeling and Wikipedia API to get images.
- Developed a web application to create rich ebooks for kids using pictures, video, and text books. The application was used to create picture books for kids and shared with parents.
Technologies: AlchemyAPI, ASP.NET, Machine Learning, JavaScript, SQL, Statistical Learning, Algorithms, Service-oriented Architecture (SOA), APIs, Architecture, Sentiment Analysis, Systems, Semantic Analysis, REST APIs, Technical Hiring, Source Code Review, Code Review, Task Analysis, Interviewing, Team Management, Research, Full-stack Development, iPaaS, API Design, Error Handling, Testing, API Integration, Technical Architecture, Software Engineering
Senior Technical Program Manager
2000 - 2009
Microsoft
- Built BizTalk Server, an enterprise messaging and workflow platform from idea to product. Released three versions of BizTalk Server.
- Created the first .NET based Outlook API, and released two versions of it.
- Built a service delivery platform for mobile telecom providers using .NET and SOA, including several WS- standards like WS-reliability and WS-eventing.
- Defined and led the inclusion of the REST API in .NET WCF.
Technologies: Windows Communication Foundation (WCF), Service-oriented Architecture (SOA), Outlook, BizTalk Server, .NET, JSON REST APIs, SDKs, SQL, REST, C++, Algorithms, Windows, ASP.NET, C#, APIs, Architecture, Systems, Workflow, RSS Feeds, Agile, REST APIs, ETL, Technical Hiring, Task Analysis, Interviewing, Team Management, Research, API Design, Error Handling, Testing, API Integration, Communication, Data Pipelines, Documentation, Technical Architecture
Technical Product and Program Manager
2006 - 2007
Amazon
- Captured the end-to-end Amazon retail messaging and workflow blueprint by collaborating with over 40 teams at Amazon that used the messaging and workflow framework.
- Defined and led the creation of an internal distributed configuration store modeled on DNS.
- Proposed a well-received vision for (the then new) Amazon cloud. It was based on true elasticity and automatic scalability.
- Presented the proposal to a special future architecture group, acting upon the vice president's suggestion. They incorporated it into their plans.
- Drove the adoption of a distributed configuration store across teams in the company.
Technologies: Java, Workflow, Engineering, SDKs, SaaS, Amazon Web Services (AWS), RESTful Development, REST APIs, SQL, C++, Algorithms, APIs, Architecture, Technical Hiring, Task Analysis, Interviewing, API Design, Error Handling, Testing, API Integration, Communication, Data Pipelines, Documentation, Technical Architecture
Senior Software Engineer
1998 - 2000
Microsoft (via Aditi)
- Built a code profiler for Visual Studio Internal tools in C++.
- Developed an XSD (XML Schema Definition) library in C++.
- Created a persistence layer for an XSLT based BizTalk schema mapper.
- Ported the FrontPage server extensions to a pre-release version of C#.
Technologies: Profiling, CODE, Visual Studio, XSLT, XSD, XML, ATL, C++, SQL, Algorithms, Windows, ASP.NET, .NET, C#, APIs, Architecture, Systems, Technical Hiring, Source Code Review, Code Review, Task Analysis, Interviewing, Team Management, API Design, Error Handling, Testing, API Integration, Communication, Data Pipelines, Documentation, Technical Architecture, Software Engineering
Experience
Maibook
https://maibook.appA private community of personalized AI agents expanding on your every activity, on-device on Mac or Windows. Local-first by design - your data never leaves your computer. “The second brain is an AI snapshot of you. The third brain is an ongoing AI live-streaming feed of your activity.” Free to try.
Desktop, macOS, Windows, Local AI, MLAI LLC.
Desktop, macOS, Windows, Local AI, MLAI LLC.
Claude Code CLI Viewer
https://rcanand.gumroad.com/l/ccviewerClaude Code Viewer is a single HTML file that transforms your local Claude Code history into a beautiful, browsable interface.
• One HTML file - No installation, no dependencies, no server
• Dark mode interface
• Full conversation history
• See Claude's thinking - Expand thinking blocks to understand the reasoning (Claude doesn't show this anymore).
• Tool visualization - See exactly how Claude uses Read, Write, Edit, Bash, and other tools
• Syntax highlighting
• Diff views - See file edits with red/green highlighting
• Search and filter
• Export to Markdown
• One HTML file - No installation, no dependencies, no server
• Dark mode interface
• Full conversation history
• See Claude's thinking - Expand thinking blocks to understand the reasoning (Claude doesn't show this anymore).
• Tool visualization - See exactly how Claude uses Read, Write, Edit, Bash, and other tools
• Syntax highlighting
• Diff views - See file edits with red/green highlighting
• Search and filter
• Export to Markdown
OllamaDash - live AI model intelligence dashboard ·
https://ollamadash.up.railway.app/home/• Python (FastAPI), vanilla JS, Railway
• Built and shipped a dashboard aggregating **20,000+ LLMs across 4 ecosystems** (Hugging Face, Ollama, OpenRouter, Artificial Analysis), with unified cross-source search, multi-key sorting, and rich faceted filtering (quantization, context window, license, parameter count, capabilities).
• Designed a **price-performance "value" metric** that fits the market's score-vs-log-price exchange rate via **per-benchmark OLS regression**, yielding a self-recalibrating "beats-the-market" residual; documented the statistical rationale and validated it with a Python mirror implementation tested against live data.
• Implemented an interactive **SVG radar chart** (zero dependencies) comparing models across 15 benchmarks (MMLU Pro, GPQA, AIME, Terminal-Bench…) in raw or price-adjusted mode, with shareable URL-encoded state.
• Engineered resilient data pipelines: background daemons refreshing every 12h, incremental diff-based scraping, schema validation, and atomic writes; responsive UI verified with **Playwright** across phone/tablet/desktop.
• AI Assisted coding - Pi.dev, Minimax-M3, Claude Code
• Built and shipped a dashboard aggregating **20,000+ LLMs across 4 ecosystems** (Hugging Face, Ollama, OpenRouter, Artificial Analysis), with unified cross-source search, multi-key sorting, and rich faceted filtering (quantization, context window, license, parameter count, capabilities).
• Designed a **price-performance "value" metric** that fits the market's score-vs-log-price exchange rate via **per-benchmark OLS regression**, yielding a self-recalibrating "beats-the-market" residual; documented the statistical rationale and validated it with a Python mirror implementation tested against live data.
• Implemented an interactive **SVG radar chart** (zero dependencies) comparing models across 15 benchmarks (MMLU Pro, GPQA, AIME, Terminal-Bench…) in raw or price-adjusted mode, with shareable URL-encoded state.
• Engineered resilient data pipelines: background daemons refreshing every 12h, incremental diff-based scraping, schema validation, and atomic writes; responsive UI verified with **Playwright** across phone/tablet/desktop.
• AI Assisted coding - Pi.dev, Minimax-M3, Claude Code
AutoML
This is an automated machine learning framework in Microsoft Azure. I was the data scientist on this team, improving the automated machine learning performance by running ML on a large corpus of benchmark datasets, and identifying ways to improve our ML pipeline.
Visual Guides
https://www.amazon.com/author/rcanandAn automated Agentic LLM and image generation AI model pipeline and engine that creates visual guides based on given topics. I evaluated the launch on Gumroad, the web, and the iOS App Store and finally settled on launching them as Kindle books.
Ahai - Local AI Idea Discovery
https://ahai.app/Native Mac desktop app using Apple MLX to scan markdown files and surface forward-looking concepts - features to build, content to create, ideas to explore. “Rediscover what you meant to do.” 100% local, privacy-first.
AI Apps Explorer
https://gisthost.github.io/?7023b76dde12cbb8f37deccfb122d49d/ai-apps-explorer.htmlAI Apps Explorer is a single-page, dependency-free catalog of 1,200+ AI tools that lets you search, filter, and sort the landscape at a glance. It
covers 17 categories spanning AI code editors, IDE extensions, coding CLI agents, code generation, deep research, search engines, browsers, browser
extensions and automation, computer-use agents, local LLM runners, writing and productivity, note-taking and knowledge tools, workflow automation,
terminals and shells, and domain-specific research. Refine further by form factor, license (open source vs proprietary), and deployment (cloud vs
local-only). Each entry captures the tool's name, form factor, license, deployment model, and a concise plain-English description of what it does, and the list updates live as you toggle filters. A built-in prompt generator turns the current filter state into a ready-to-paste natural-language summary you can copy straight into any LLM. Built as one self-contained HTML file with no install, no server; runs anywhere by opening the link.
covers 17 categories spanning AI code editors, IDE extensions, coding CLI agents, code generation, deep research, search engines, browsers, browser
extensions and automation, computer-use agents, local LLM runners, writing and productivity, note-taking and knowledge tools, workflow automation,
terminals and shells, and domain-specific research. Refine further by form factor, license (open source vs proprietary), and deployment (cloud vs
local-only). Each entry captures the tool's name, form factor, license, deployment model, and a concise plain-English description of what it does, and the list updates live as you toggle filters. A built-in prompt generator turns the current filter state into a ready-to-paste natural-language summary you can copy straight into any LLM. Built as one self-contained HTML file with no install, no server; runs anywhere by opening the link.
Maiweb - Personal Content Aggregator
Designed, built, and operated a production news reader aggregating 1,000+ RSS/API sources into a 120,000+ item corpus, serving a trending-ranked, keyboard-driven feed.
• Designed a transparent trending algorithm (TF-IDF bigram scoring with curated stopword layers, source-diversity thresholds, deterministic ranking, cross-source dedup) instead of an opaque ML black box - every ranking decision is explainable and tunable via config.
• Built a polite ingestion pipeline: per-domain rate limiting, ETag-based HTTP caching, jittered scheduling, and self-healing error handling that auto-quarantines broken feeds.
• Architected a zero-user-state back end: all read/save/hide personalization lives in versioned browser localStorage, eliminating auth, per-user tables, and privacy surface area.
• Practiced spec-driven AI-assisted development: wrote file-by-file implementation specs and agent guardrail docs (AGENTS.md) to direct AI coding agents safely against a live production system, with pytest regression coverage for ranking heuristics.
• Enforced content and licensing compliance with automated paywall detection and a dependency license allow-list audit.
• Designed a transparent trending algorithm (TF-IDF bigram scoring with curated stopword layers, source-diversity thresholds, deterministic ranking, cross-source dedup) instead of an opaque ML black box - every ranking decision is explainable and tunable via config.
• Built a polite ingestion pipeline: per-domain rate limiting, ETag-based HTTP caching, jittered scheduling, and self-healing error handling that auto-quarantines broken feeds.
• Architected a zero-user-state back end: all read/save/hide personalization lives in versioned browser localStorage, eliminating auth, per-user tables, and privacy surface area.
• Practiced spec-driven AI-assisted development: wrote file-by-file implementation specs and agent guardrail docs (AGENTS.md) to direct AI coding agents safely against a live production system, with pytest regression coverage for ranking heuristics.
• Enforced content and licensing compliance with automated paywall detection and a dependency license allow-list audit.
Obsidian Pencil - Infinite Whiteboard for Handwriting with Apple Pencil
https://community.obsidian.md/plugins/pencilAn infinite whiteboard for Obsidian. Draw and handwrite with a stylus, mouse, or finger — saved as .pencil files right inside your vault.
Pencil is an opinionated, single-slice take on handwriting in Obsidian. The existing whiteboard apps are heavy and loaded with features most people never touch. Pencil does one thing well: a fast, infinite canvas that feels like pen on paper, and nothing more.
Infinite canvas — pan and zoom endlessly; your view position is saved with the note.
Stylus-first, pressure-aware — Apple Pencil and other pens get real pressure-thickness variation. Toggle it off anytime.
Mouse and finger, too — no stylus? Draw with a mouse or finger. Pressure is simply held constant.
Palm rejection — once you've used a pen, finger touches become panning, so you can rest your hand.
Eraser — stroke-level erase by dragging over what you don't want.
Select and move — box-select strokes and drag them around.
Colors and sizes — 8 built-in colors plus your own custom palette. Four stroke widths.
Saved as vault files — each whiteboard is a .pencil JSON file, versioned and synced with the rest of your notes.
Works everywhere — desktop, iPad, and mobile. The toolbar adapts: icons on desktop, short text labels on mobile.
Pencil is an opinionated, single-slice take on handwriting in Obsidian. The existing whiteboard apps are heavy and loaded with features most people never touch. Pencil does one thing well: a fast, infinite canvas that feels like pen on paper, and nothing more.
Infinite canvas — pan and zoom endlessly; your view position is saved with the note.
Stylus-first, pressure-aware — Apple Pencil and other pens get real pressure-thickness variation. Toggle it off anytime.
Mouse and finger, too — no stylus? Draw with a mouse or finger. Pressure is simply held constant.
Palm rejection — once you've used a pen, finger touches become panning, so you can rest your hand.
Eraser — stroke-level erase by dragging over what you don't want.
Select and move — box-select strokes and drag them around.
Colors and sizes — 8 built-in colors plus your own custom palette. Four stroke widths.
Saved as vault files — each whiteboard is a .pencil JSON file, versioned and synced with the rest of your notes.
Works everywhere — desktop, iPad, and mobile. The toolbar adapts: icons on desktop, short text labels on mobile.
One File - Edit Markdown File from Anywhere Inside Your Obsidian Vault
https://community.obsidian.md/plugins/one-fileEdit any markdown file on your computer from inside an Obsidian vault, in a distraction-free focus mode.
Instead of copying files into the vault, One File creates a link to the original file on disk. Edits in Obsidian write through to the real file — no duplicates, no sync logic.
Desktop only (macOS / Windows / Linux). The plugin cannot be installed or run on mobile — Obsidian will not show it in the community plugin browser on mobile, and it will not load if installed manually. See the Mobile caveat below for how to work with your files on mobile.
github repo - https://github.com/rcanand/obsidian-one-file
Instead of copying files into the vault, One File creates a link to the original file on disk. Edits in Obsidian write through to the real file — no duplicates, no sync logic.
Desktop only (macOS / Windows / Linux). The plugin cannot be installed or run on mobile — Obsidian will not show it in the community plugin browser on mobile, and it will not load if installed manually. See the Mobile caveat below for how to work with your files on mobile.
github repo - https://github.com/rcanand/obsidian-one-file
NotebookLM Downloader
https://github.com/rcanand/nblm_dlSolves the problem of downloading a research report from my NotebookLM notebooks.
# NotebookLM Report Downloader (Chrome extension)
A minimal MV3 extension that downloads the **Source guide** and the **Report** from a report source you have open in NotebookLM (notebooklm.google.com), as two Markdown files.
It only reads content that has already been rendered in your own signed-in browser tab. It does not make any extra network requests to Google beyond what the page itself already does, and it never touches data that is not visible to you.
## What it captures
When you open an imported report source in NotebookLM, the page renders a Angular component with two sections:
1. **Source guide** — a fixed summary paragraph + the "Key topics" chips. Saved as `source.md`.
2. **Report** — the scrollable document body (headings, paragraphs, bulleted lists, tables, inline bold/italics/links, and citation markers like `[1, 2]`). Saved as `report.md`.
# NotebookLM Report Downloader (Chrome extension)
A minimal MV3 extension that downloads the **Source guide** and the **Report** from a report source you have open in NotebookLM (notebooklm.google.com), as two Markdown files.
It only reads content that has already been rendered in your own signed-in browser tab. It does not make any extra network requests to Google beyond what the page itself already does, and it never touches data that is not visible to you.
## What it captures
When you open an imported report source in NotebookLM, the page renders a Angular component with two sections:
1. **Source guide** — a fixed summary paragraph + the "Key topics" chips. Saved as `source.md`.
2. **Report** — the scrollable document body (headings, paragraphs, bulleted lists, tables, inline bold/italics/links, and citation markers like `[1, 2]`). Saved as `report.md`.
Asta Report Viewer
https://gisthost.github.io/?e647e0142cff6e817deafe5eed73b2fb/asta_viewer.htmlBuilt a viewer to navigate Allen AI Asta research reports, exported as JSON.
I needed it for myself, so I vibe-coded it using Claude Code.
Allen AI Asta has fantastic, research-backed deep reports on any topic. I wanted to manage my favorite research reports offline, and had no way to do that directly with Asta. Hence, built it for myself.
I needed it for myself, so I vibe-coded it using Claude Code.
Allen AI Asta has fantastic, research-backed deep reports on any topic. I wanted to manage my favorite research reports offline, and had no way to do that directly with Asta. Hence, built it for myself.
Ganglion
A web application that empowers consumers to manage their news consumption experience providing full control and complete privacy.
I conceived, designed, built, and deployed this project end-to-end and automated the daily refresh of news content, updating machine learning models and word cloud images of the most common news topics. This set up the entire app to work on autopilot.
I optimized the performance to efficiently fetch approximately 5,000 distinct news articles every day from 400 news feeds and manually labeled data to train a model for clickbait detection in news articles with 83% accuracy. I trained and updated models to detect sentiment, objectivity, infer which articles have clickbait headings, perform NER, and generate word clouds. I fetched and parsed news articles daily, updated and generated models and document embeddings, created word cloud images, and uploaded to AWS S3 runs within hours.
I conceived, designed, built, and deployed this project end-to-end and automated the daily refresh of news content, updating machine learning models and word cloud images of the most common news topics. This set up the entire app to work on autopilot.
I optimized the performance to efficiently fetch approximately 5,000 distinct news articles every day from 400 news feeds and manually labeled data to train a model for clickbait detection in news articles with 83% accuracy. I trained and updated models to detect sentiment, objectivity, infer which articles have clickbait headings, perform NER, and generate word clouds. I fetched and parsed news articles daily, updated and generated models and document embeddings, created word cloud images, and uploaded to AWS S3 runs within hours.
Meon - Web Platform That Creates Apps in Minutes
https://youtu.be/ZCM7V_QH1zkThis is a Ruby on Rails-based web app that creates a wide variety of apps and starts using them immediately-in minutes. I conceived, designed, and built this product entirely on my own. I created 85 apps from a wide range of domains on this platform in just two days. Using this product, I built business apps for various early clients. The video below shows a tour of the variety of apps created in Meon, most of them in less than five minutes.
An AI Web App
An AI web app with a complex data model designed and built for a client of my employer. It was built using Python, Django, Django REST Framework, TypeScript, and Angular; it was deployed to Google Cloud Platform (GCP).
Sumurai
Sumurai is a math puzzle game conceived, designed, and implemented by me in 2016. It enabled young kids to learn arithmetic by playing a challenging puzzle game. I collaborated with researchers, educators, and customers to improve the game. It received great feedback from researchers and teachers in the US who used it with their students.
Smart Run
An iOS educational game I designed, built, launched, and marketed. It merged an educational experience with a fast-paced running game. It was designed to add more educational content without modifying the core game code.
Education
1990 - 1994
Bachelor's Degree in Engineering
Indian Institute of Technology - Roorkee, India
Certifications
MAY 2022 - PRESENT
Fundamentals of Reinforcement Learning
University of Alberta | via Coursera
APRIL 2022 - PRESENT
TensorFlow Developer Certificate
DeepLearning.AI
APRIL 2022 - PRESENT
DeepLearning.AI TensorFlow Certification
Coursera
JANUARY 2022 - PRESENT
Cryptocurrency Forecasting Using Machine Learning in Power BI
Coursera
FEBRUARY 2021 - PRESENT
Discrete Mathematics and Analyzing Social Graphs
Higher School of Economics, National Research University | via Coursera
JUNE 2020 - PRESENT
Natural Language Processing with Classification and Vector Spaces
DeepLearning.AI | via Coursera
JUNE 2020 - PRESENT
Natural Language Processing with Probabilistic Models
DeepLearning.AI | via Coursera
MAY 2020 - PRESENT
Mathematics for Machine Learning: Linear Algebra
Imperial College London | via Coursera
APRIL 2020 - PRESENT
Data Structures
UC San Diego | via Coursera
APRIL 2020 - PRESENT
Data Structures
UC San Diego and HSE | via Coursera
MARCH 2020 - PRESENT
Algorithmic Toolbox
UC San Diego | via Coursera
MARCH 2020 - PRESENT
Algorithmic Toolbox
UC San Diego and HSE | via Coursera
MAY 2018 - PRESENT
Improving Deep Neural Networks: Hyperparameter Tuning, Regularization and Optimization
DeepLearning.AI | via Coursera
MAY 2018 - PRESENT
Sequence Models
DeepLearning.AI | via Coursera
MAY 2018 - PRESENT
Convolutional Neural Networks
DeepLearning.AI | via Coursera
MAY 2018 - PRESENT
Neural Networks and Deep Learning
DeepLearning.AI | via Coursera
MAY 2018 - PRESENT
Structuring Machine Learning Projects
DeepLearning.AI | via Coursera
MAY 2018 - PRESENT
Deep Learning Specialization (Six courses)
DeepLearning.AI | via Coursera
AUGUST 2016 - PRESENT
Statistical Learning
Stanford Online
NOVEMBER 2015 - PRESENT
Introduction to Mathematical Thinking
Stanford | via Coursera
JUNE 2013 - PRESENT
Human Computer Interaction
UC San Diego | via Coursera
Skills
Libraries/APIs
REST APIs, Pandas, Matplotlib, Natural Language Toolkit (NLTK), SpaCy, NumPy, Scikit-learn, Fast.ai, PyTorch, TensorFlow, OpenAI Assistants API, Claude API, SciPy, ATL, LSTM, OpenCV, React, jQuery, Django ORM, Keras, Tidyverse, Ggplot2, Azure Cognitive Services, Python API, ReportLab, Pydantic, PySide, OpenAPI
Tools
AutoML, ChatGPT, PaLM 2, Claude, Gecko, Azure Machine Learning, Gensim, Seaborn, Microsoft Power BI, Jupyter, Visual Studio, JSX, Named-entity Recognition (NER), Trello, H2O AutoML, Ansible, Looker, Amazon SageMaker, GPT Builder, Claude Code, AI Prompts
Languages
Ruby, SQL, Python 3, Python, JavaScript, Snowflake, Objective-C, C#, Java, XML, XSD, XSLT, TypeScript, Lua, R, C++, Markdown
Frameworks
Ruby on Rails (RoR), Flask, Agentic Frameworks, Corona SDK, ASP.NET, .NET, CODE, Bootstrap, Angular, AngularJS, Unity, Cocos3d, Django, Django REST Framework, Tauri, Qt, FastMCP
Paradigms
REST, Agile, RESTful Development, Testing, ETL, Model Context Protocol (MCP), Service-oriented Architecture (SOA), App Store Optimization (ASO), Kanban, Foundation Models
Storage
Data Pipelines, PostgreSQL, MongoDB
Platforms
Google Cloud Platform (GCP), Visual Studio Code (VS Code), Amazon Web Services (AWS), Azure IaaS, Linux, Windows, MacOS, Heroku, AlchemyAPI, iOS, Docker, Azure, Jupyter Notebook, Ollama, Open WebUI
Industry Expertise
Healthcare
Other
Natural Language Processing (NLP), Neural Networks, Algorithms, SaaS, Computer Vision, Machine Learning, Artificial Intelligence (AI), Exploratory Data Analysis, APIs, Data Science, Technical Hiring, Source Code Review, Code Review, Task Analysis, Interviewing, Generative Pre-trained Transformers (GPT), Language Models, Chatbots, Chatbot Conversation Design, Large Language Models (LLMs), Retrieval-augmented Generation (RAG), OpenAI, OpenAI GPT-4 API, OpenAI GPT-3 API, Google, LangChain, AI Research, Benchmarking, Datasets, Evaluation, Vector Stores, Vector Search, Prompt Engineering, AI Agents, MTEB, Large Model Systems Organization (LMSYS), Google Colaboratory (Colab), Generative Artificial Intelligence (GenAI), Generative Pre-trained Transformer 3 (GPT-3), Embedding Models, Open-source LLMs, Agentic AI, Data Analysis, AI Chatbots, Data Engineering, Optical Character Recognition (OCR), API Design, Error Handling, Remote Work, API Integration, Communication, LLM Reasoning, Vector Databases, Documentation, Anthropic, Conversational AI, Small Language Models (SLMs), Software Engineering, Architecture, Artificial Neural Networks (ANN), Image Recognition, Sentiment Analysis, JSON REST APIs, SDKs, Deep Learning, Statistical Learning, Web Scraping, 3D Image Processing, Image Processing, Team Management, ChromaDB, Qdrant, Research, Full-stack Development, iPaaS, Meta Llama, RAG Architecture, FastAPI, Open Source, Image Generation, Multimodal GenAI, Technical Architecture, Laspy, PDAL, BizTalk Server, Outlook, Windows Communication Foundation (WCF), Systems, Engineering, Workflow, Profiling, Convolutional Neural Networks (CNNs), Deep Neural Networks (DNNs), Data Structures, Recurrent Neural Networks (RNNs), Gated Recurrent Unit (GRU), Sequence Models, Hyperparameters, Regularization, Semantic Analysis, Rankings, Tf-idf, RSS Feeds, Discrete Mathematics, Graph Theory, Games, 2D Games, Mathematics, Game Art, LiDAR, Full-stack, Real-time Data, Interactive Voice Response (IVR), Dialog Systems, Machine Learning Operations (MLOps), Google Speech-to-Text, Labeling, Time Series Analysis, Reinforcement Learning, Time Series, Fintech, Cryptocurrency, Medical Imaging, Hybrid Search, Amazon Bedrock AgentCore, Llama 2, GitHub Copilot Chat, Hugging Face, Stable Diffusion, Cursor AI, Qwen, AI2 models, Quantization, AI2 Models, Hugging Face Diffusers, Artificial General Intelligence (AGI), Multistage LLM Chains, ChatGPT API, PDF, Conversational Agent, Pi.dev, LiteLLM, uv, OpenRouter, Obsidian, Agentic Coding, Agentic AI Systems, Agentic RAG Systems, Agentic Workflow Design, Vibe Coding, Obsidian plugins, Browser Plugins, AI Tools, LanceDB, Software Architecture, RAG Pipelines, RAG Systems, MLX, Pi dev, AI assisted coding, Metrics, AI-generated Code, UX Engineering, UI Plugins, Context Engineering, Loop Engineering, Harness Engineering, Pi Dev, Model Evaluation, Nanocode, Z Image Turbo, Blackforest Labs Flux, Automatic1111, RESTFul APIs, Obsidian Plugins, Web Search, Desktop App, Fine-tuning, Text-to-Speech (TTS)
Collaboration That Works
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
1
Share your needs
Discuss your requirements and refine your scope in a call with a Toptal domain expert.
2
Choose your talent
Get a short list of expertly matched talent within 24 hours to review, interview, and choose from.
3
Start your risk-free talent trial
Work with your chosen talent on a trial basis for up to two weeks. Pay only if you decide to hire them.
Top talent is in high demand.
Start hiring