Purshottam S.
FDE | LLMO - RAG - Agentic Workflows
Most AI projects fail not because of bad code — but because nobody connected the technical decisions to the business goals. That's the gap I close. I'm a Full Stack Architect and Technical PM with 14+ years of engineering experience, an IIM Indore management credential, and hands-on expertise in LLM systems, RAG pipelines, and agentic workflows. I've delivered solutions across fintech, banking, insurance, automotive, retail, and AI SaaS — working with teams of 1 to 20 engineers, managing budgets, writing Scopes of Work, and shipping production systems that keep running long after the contract ends. ── What makes me different ── I operate at two levels most freelancers can't: → Product level: I define what needs to be built, write the technical roadmap, scope the project honestly, and align the architecture with your business constraints — before a single line of code is written. → Engineering level: I build it. LLM integrations, RAG pipelines with semantic search, multi-agent systems, APIs, cloud infrastructure, CI/CD — end to end. This means you don't need a separate PM and a separate architect. You get both, with no translation layer in between. ── What I build with AI ── → RAG Systems: AI that answers from your actual data — documents, databases, internal knowledge — accurately and at scale. Built with LangChain, LlamaIndex, Pinecone, pgvector, and Weaviate. → Agentic Workflows: Multi-agent systems using LangGraph and CrewAI that automate complex, multi-step business processes — with the guardrails to keep them reliable in production. → LLM Observability: I don't just ship AI — I instrument it. RAGAS evaluations, LangSmith tracing, and performance benchmarking so you know what's working and what isn't. → AI on AWS / Azure / GCP: Bedrock, Lambda, serverless — secure, scalable, cost-optimized deployments. ── Project and Delivery ── Technical roadmap development · Scopes of Work · Agile / Scrum / Kanban · Budget management · Team hiring and mentoring · Jira, Confluence, Monday ── Core engineering stack ── Python (FastAPI, Django) · Node.js · React · Next.js · TypeScript · PostgreSQL · MongoDB · Docker · Kubernetes · CI/CD · OAuth2 / JWT ── Industries ── Fintech · Banking · Insurance · Automotive · Retail · AI SaaS · Media I'm direct and proactive. If scope needs to shift, if a simpler solution saves you three months, or if something is going off track — you'll hear it from me before it becomes a problem. If you're building an AI product and need someone who thinks at the business level and executes at the engineering level — let's talk.
Earnings and projects
Last 6 months ÷ 6
Last 6 months: $4.8k
Billed in all: $94,377$94k
$1.6k in the last 6 months
Across 18 contracts
Earnings, last 30 days
Skill rankings: Top 12% in Blockchain Architecture
Ranked by MRR among everyone on Upwork with the skill.
- #36Blockchain Architecture305 freelancers · 🇮🇳 #14Top 12%🇮🇳#14 of 92
- #193Solution Architecture1,009 freelancers · 🇮🇳 #24Top 19%🇮🇳#24 of 207
- #335Technical Project Management1,684 freelancers · 🇮🇳 #21Top 20%🇮🇳#21 of 145
- #410Product Management2,632 freelancers · 🇮🇳 #30Top 16%🇮🇳#30 of 288
- #441Google Cloud Platform2,939 freelancers · 🇮🇳 #70Top 15%🇮🇳#70 of 591
- #627Large Language Model2,726 freelancers · 🇮🇳 #136Top 23%🇮🇳#136 of 581
- #891Retrieval Augmented Generation3,618 freelancers · 🇮🇳 #193Top 25%🇮🇳#193 of 801
- #1,051Generative AI5,038 freelancers · 🇮🇳 #255Top 21%🇮🇳#255 of 1,206