Skip to content
Hey, I'm Ashok

Available for freelance & full-time roles

Hey, I'm Ashok

AI Engineer & Full-Stack Developer

I turn ideas into products, with AI baked in and the polish to match. Full-stack developer, AI engineer, and someone who really likes shipping. Got something in mind? Let's make it real.

View WorkContact Me
#AK-001Prod Ready

Hey, I'm Ashok

AI Engineer · Full-Stack Dev

  • AI
  • LLMs
  • RAG
  • Agents
  • EVAL
  • FULL-STACK

Move to tilt · tap ↻ to flip

About

From idea to production.

I'm Ashok, a full-stack developer and AI engineer with about 8 years in the industry. I've shipped a lot of products in that time, and the biggest lesson has stayed the same: good software comes from understanding people first and code second. Before I write a line, I want to know who it's for and what problem it really solves.

When we work together, I treat your product like it's my own. I ask a lot of questions early, share progress often, and tell you honestly when something isn't working, so we can fix it before it costs you. I enjoy the whole journey, from a rough idea on a call to the moment real users get their hands on it. If you've got an idea, let's connect and talk it out.

  • People first

    I start with the people who'll use the product, then pick the tech that serves them best.

  • Ownership

    I care about what happens after launch as much as before it. If I built it, I stand behind it.

  • Honest and clear

    Regular check-ins, plain language, and no surprises. You'll always know where things stand.

Projects

Selected work.

Skills

Tools I ship with.

  • LLMs & GenAI

    • OpenAI
    • Anthropic Claude
    • Google Gemini
    • Open-source LLMs (Llama, Mistral, Qwen)
    • Prompt Engineering
    • Structured Outputs
    • Function / Tool Calling
    • Streaming Responses
    • Multimodal (Vision & Audio)
    • Context Engineering
    • Guardrails & Safety
    • Token & Cost Optimization

  • RAG & Retrieval

    • Retrieval-Augmented Generation
    • Embeddings
    • Chunking Strategies
    • Hybrid Search (BM25 + Vector)
    • Re-ranking
    • Query Rewriting
    • Knowledge Graphs / GraphRAG
    • Document Parsing & OCR
    • Citations & Source Grounding
    • pgvector
    • Pinecone
    • Weaviate
    • Qdrant
    • Chroma
    • Elasticsearch

  • AI Agents

    • Agentic Workflows
    • Multi-Agent Systems
    • LangChain
    • LangGraph
    • LlamaIndex
    • Vercel AI SDK
    • OpenAI Agents SDK
    • MCP (Model Context Protocol)
    • Memory & State Management
    • Human-in-the-Loop

  • Evals & LLMOps

    • LLM Evaluation
    • RAG Evaluation
    • Golden Datasets
    • Ragas
    • DeepEval
    • Hallucination Detection

  • ML & Model Training

    • Python
    • PyTorch
    • Hugging Face
    • Fine-tuning (LoRA, QLoRA)
    • Model Distillation
    • vLLM
    • Ollama
    • Model Quantization
    • scikit-learn
    • Pandas & NumPy

  • Frontend

    • TypeScript
    • JavaScript
    • React
    • Next.js
    • Tailwind CSS
    • shadcn/ui
    • Framer Motion
    • Three.js
    • Redux / Zustand
    • TanStack Query
    • Accessibility

  • Backend

    • Node.js
    • Express
    • NestJS
    • Python
    • REST APIs
    • GraphQL
    • WebSockets
    • gRPC

  • Cloud & Shipping

    • AWS
    • Vercel
    • Docker
    • Kubernetes
    • CI/CD
    • GitHub Actions
    • Serverless
    • Monitoring & Logging

Contact

Let's build something real.

Every product I've shipped started as a simple conversation. If you have an idea, even a rough one, tell me about it. I'll listen, ask a few good questions, and give you an honest take on how we could make it real. Worst case, you walk away with a clearer plan.