Jyothi Kumar Dummala
Software Engineer / AI Engineer
Building agents that close deals, not just demo well. Shipping in Research, Sales, GTM domains.
I build production-grade GenAI products end-to-end - provider orchestration, prompting + context engineering, RAG, fine-tuning, agentic workflows, and evaluation frameworks. I love to understand the domain, map the problem to a reliable solution and scale it!
I have experience in shipping agents to sales, market research and GTM domains. Where I map the problems, ship agents (diff architecture agents) and then deliver the solution!
Ownership, collaboration, and measurable business outcomes.
Multi-LLM integrations
OpenAI, Anthropic, DeepSeek, Gemini with project-specific evaluation frameworks.
90%+ accuracy lift
Advanced prompting, context management, RAG, and fine-tuning strategies.
Long-running agents
500+ tool calls with 90% accuracy via rigorous context engineering.
Unified LLM Wrapper
Agentic wrapper with retries/fallbacks and 500+ downloads.
Voice-first AI tutor
Hands-free Spanish tutor with real-time STT → LLM → TTS pipeline and 5 teaching modes.
Experience
ProdGain
Software Engineer (AI / Full-stack)
- Built full-stack MERN applications with integrated GenAI capabilities.
- Integrated OpenAI, Anthropic, DeepSeek, Gemini and built evaluation frameworks.
- Improved LLM performance by 70%+ using prompting, context, RAG, fine-tuning.
- Architected long-running agents with 500+ tool calls and 90% accuracy.
- Owned product scoping, roadmaps, and milestone delivery for client builds.
- Led design reviews and shipped UI, backend, and AI workflows end-to-end.
- Instrumented reliability checks, eval suites, and release checklists.
- Took end-to-end ownership and supported hiring, knowledge transfer, and architecture reviews.
ADP
Software Developer Intern
- Built backend CRUD routes in Node.js.
- Automated SQL queries and API calls; improved code structure with JS, jQuery, SQL, HTML, CSS.
- Collaborated with QA and product to deliver internal tooling updates.
- Refactored legacy modules for maintainability and clearer data flow.
Brainlox
Full Stack Engineer (Remote)
- Developed React.js UI pages and integrated with AWS services.
- Translated design requirements into responsive UI components.
- Shipped features with cloud integrations and clean release notes.
KMIT - Teleparadigm Networks
R&D Intern
- Built an event-driven drone control app with a custom Alexa skill.
- Worked with AWS API Gateway websockets, Lambda, DynamoDB, and drone programming.
- Prototyped event flows, telemetry, and interactive demos.
- Documented architecture and findings for research review.
Recent work
Aarogya — Medical Customer Support
AI chatbot for a health insurance company with a custom knowledge-base, end-to-end ticket handling with real-time email notifications, and web-search grounded to admin-curated domains.
Research Paper Judge
Multi-agent AI system that evaluates arXiv research papers and generates peer-review-style reports with PASS/FAIL verdicts and scored dimensions.
Unified LLM Wrapper
A developer-friendly agentic wrapper across providers with reasoning, retries, and fallbacks.
Voice Tutor — Spanish
Voice-first AI agent for learning Spanish through natural conversation — real-time STT, LLM orchestration, TTS, 5 teaching modes, and persistent sessions.
Skills
AI / GenAI
Web Development
Programming
Databases
Cloud / DevOps
My 5-tier GenAI build approach
Provider + orchestration
Route across OpenAI, Anthropic, Gemini, and DeepSeek with unified interfaces, automatic retries, and graceful fallbacks — no glue code, no provider lock-in.
Prompting + context engineering
Shape what the model sees: structured payloads, trust tiers, freshness policies, and constraint schemas that make output predictable before touching a single prompt word.
RAG + tool use
Precision retrieval with intent-aware queries, metadata filtering, and source-grounded output — tools with typed contracts and bounded retries so failures stay explicit.
Fine-tuning + adaptation
When prompting and retrieval hit their ceiling, fine-tune on domain-specific data to push accuracy, consistency, and latency beyond what off-the-shelf models can deliver.
Agentic systems + evaluation
Long-running agents with structured plans, 500+ tool call budgets, and eval frameworks that measure real outcomes — not just whether the output looks good.
Blog
Reliability-first LLM systems: what actually works in production
February 14, 2026 · 8 min read
How I design evals, fallbacks, and release guardrails for systems that survive real traffic.
Context engineering playbook I use before touching prompts
February 14, 2026 · 7 min read
The practical method I follow to shape inputs, retrieval, and constraints so outputs stay stable.
Contact
Let's explore ideas
Always up for interesting problems, early-stage ideas, and collaborations worth building.
Whether it's a half-formed idea or something almost ready to ship — I'm happy to think through it together.
Contact