
Closed
Posted
Paid on delivery
I’m kicking off a proof-of-concept around Google’s new Gemini multimodal model and need an experienced hand to get the fundamentals right. Your task is to spin up a small yet extensible codebase that: • Authenticates through Google AI Studio • Accepts text (and ideally image) prompts • Returns structured JSON responses • Stores key metadata in Postgres I’d like the logic wrapped in a clean REST service so my in-house team can drop it straight into our existing stack. Python is preferred, but Node.js is fine if that’s where you shine. Please containerise everything—one `docker compose up` should bring the service online and fire a successful Gemini request using the API key I’ll supply. Deliverables • Dockerised repo on GitHub/GitLab • Postman (or similar) collection demonstrating sample calls • Concise setup & usage guide (Markdown) • Light unit-test suite; code must clear PEP8/ESLint Acceptance criteria • Service launches via Docker without manual tweaks • At least two demo endpoints (plain text and image-to-text) hitting Gemini successfully • Linting and tests pass cleanly If you’ve already played with Gemini Pro or Ultra, show me. Real-world examples will move you to the top of the list.
Project ID: 40763459
19 proposals
Remote project
Active 5 days ago
Set your budget and timeframe
Get paid for your work
Outline your proposal
It's free to sign up and bid on jobs
19 freelancers are bidding on average ₹18,387 INR for this job

Hi, I’ve already worked hands‑on with multimodal AI stacks (Gemini Pro + Ultra) and know the quirks of Google AI Studio authentication, JSON structuring, and containerized deployment. For your proof‑of‑concept, I’ll build a clean, Docker‑first REST service that your team can drop straight into production: Endpoints: • /text → accepts text prompts, returns structured JSON • /image → accepts image + text, returns Gemini’s multimodal response Infra: • Postgres integration for metadata logging • One‑command startup (docker compose up) • Unit tests + linting (PEP8/ESLint) baked in Deliverables: • GitHub/GitLab repo with Dockerized codebase • Postman collection for demo calls • Concise Markdown setup guide • Lightweight test suite ensuring clean CI/CD Why me? I specialize in Python (FastAPI + SQLAlchemy) but can pivot to Node.js if you prefer. I’ve containerized similar AI services before, so I know how to avoid the “works on my machine” trap. Expect a repo that feels modern, extensible, and production‑ready—not just a throwaway demo. Style: Transparent, communicative, and focused on making your in‑house team’s life easier.
₹25,000 INR in 7 days
6.5
6.5

Hi, I can build a clean, extensible Gemini-based REST service with Docker deployment, structured responses, and PostgreSQL metadata storage. My approach would be to create a modular backend where Gemini integration, API endpoints, database operations, and configuration are separated for easy future expansion. Deliverables: • Python REST API service for Gemini integration • Google AI Studio authentication setup • Text prompt endpoint • Image-to-text multimodal endpoint • Structured JSON response handling • PostgreSQL metadata storage • Docker/Docker Compose deployment • Postman collection with demo requests • Markdown setup and usage documentation • Basic unit tests and code quality checks I’ll ensure the service starts cleanly with a single Docker Compose command, handles API errors properly, and follows maintainable coding practices. I can also structure the project so your internal team can extend it later with additional Gemini features or application workflows. Best regards Ankit
₹15,000 INR in 3 days
2.9
2.9

Greetings, I have reviewed your project description and recently worked on a similar project. I believe I can help you deliver this successfully. Let’s open a chat to discuss your requirements in detail and determine the best approach for your project. Regards
₹25,000 INR in 7 days
0.0
0.0

Hello! Your Gemini REST service in Docker, ready 2 days after the award. You pay us nothing until you've seen it answer a Gemini request. Here is what you get: • Python REST service: plain-text and image-to-text endpoints, JSON replies • Postgres table for the key metadata of every call • One "docker compose up" starts it, API key kept in your own .env file • Postman collection, Markdown setup guide, unit tests, PEP8-clean code Good to know: Gemini model names change often, so we keep the model name in one setting you can change. ✅ Included: 3 small changes within 3 weeks after you approve the work, and we fix any bug of ours for free. Bonus: a /health endpoint that checks Postgres and the Gemini key. Our own demo: an AI assistant that answers from product manuals and shows its sources. Price: ₹15,000, 50% off our regular ₹30,000 while we earn our first reviews here. Should the image endpoint take an uploaded file, or an image link? P.S. Oldrich, the owner, supervises the work and stands behind every result. We use various tools and AI agents to work fast and efficiently.
₹15,000 INR in 2 days
0.0
0.0

Hello, I can build the Gemini proof-of-concept as a small, clean REST service in Python (FastAPI), fully containerised. What you get: (1) a FastAPI service that authenticates to Google AI Studio with the API key from an .env file; (2) two demo endpoints, plain text prompt and image-to-text (image upload plus prompt), both returning structured JSON using Gemini's JSON response mode with a schema; (3) PostgreSQL storage of request metadata (timestamp, model, endpoint, token usage, latency, status) via SQLAlchemy and Alembic migrations; (4) docker compose with the API and Postgres, so one `docker compose up` starts everything; (5) a Postman collection, a concise Markdown setup and usage guide, and a light pytest suite (Gemini calls mocked in tests), with ruff and black so linting passes cleanly. Price: 20000 INR fixed, delivery in 5 days after the kickoff. Because I will not have your API key until you send it, I verify the live Gemini calls with a short test key you provide, or you run the included smoke-test script on your side. I have no linkable sample of a Gemini project to show, so I suggest I deliver the text endpoint first for a quick check. Questions: which Gemini model version do you want as default, and do you need authentication on the REST service itself?
₹20,000 INR in 5 days
0.0
0.0

Hi, I can build this Gemini POC in Node.js/TypeScript with PostgreSQL and Docker Compose. I already have hands-on experience integrating Google Gemini into backend applications, including structured JSON responses, schema validation, API-key configuration, persistence, and handling provider errors such as rate limits and temporary 5xx failures. I would deliver: - REST endpoint for text prompts - REST endpoint for image-to-text / multimodal input - Gemini integration using Google’s official Gen AI SDK - validated structured JSON responses - PostgreSQL metadata persistence - Docker Compose setup for API + database - configurable Gemini model through environment variables - basic automated tests - Postman collection - clear README/setup instructions I’ll keep the POC modular and lightweight so it can be extended later without unnecessary complexity. Before starting, I would confirm the expected JSON schema, metadata to persist, and preferred image input format.
₹20,000 INR in 3 days
0.0
0.0

You need a Gemini service your team can drop in, where one docker compose up brings it online and fires a real Gemini call. My plan: 1. REST service with /generate/text and /generate/image. AI Studio key from .env, JSON response schema so replies always parse, clean rate-limit errors. 2. Postgres in compose with a health check, storing request ID, model, tokens, latency and status. A startup smoke test sends one real Gemini request. 3. Postman collection, Markdown guide, unit tests with mocked Gemini calls, lint clean. Proof: my Job Radar workflow (in my portfolio) has an LLM score hundreds of jobs a day as structured JSON, Groq first with Gemini as fallback. I have no reviews yet, so each milestone is paid only after you check it against your acceptance list. INR 17,000 fixed, 5 days. Quick question: for images, file uploads, URLs, or both? Peter Rock
₹17,000 INR in 5 days
0.0
0.0

Hi! I can build your Gemini-powered application as a clean, maintainable, Dockerized service using Python or Node.js, with PostgreSQL integration and well-documented API endpoints. My implementation plan: 1. Gemini API integration — configure secure API access, implement text and image processing, and return structured JSON responses. 2. Backend and database — develop REST endpoints, PostgreSQL models, input validation, and reliable error handling. 3. Docker setup — provide a working Docker Compose configuration with environment-based settings. 4. Testing and handover — include automated tests, a Postman collection, setup instructions, and clear API documentation. I focus on delivering complete, testable applications rather than isolated code snippets. A relevant example from my portfolio is the Small Business FAQ Chatbot Widget, which demonstrates my work with AI-powered application functionality. I can start promptly, share progress throughout development, and deliver the complete project within 5 days. One quick question: do you already have a preferred Gemini model, or should I recommend one based on your text and image processing requirements? Looking forward to working with you!
₹24,000 INR in 5 days
0.0
0.0

I will develop a clean, extensible, and Docker-based REST API for your Gemini multimodal PoC. It will authenticate using a Google AI Studio API Key to process text and image prompts, return structured JSON, and store necessary metadata in PostgreSQL. My approach: I will create two demo endpoints—text-to-JSON and image-to-text—using Python, FastAPI, the Gemini API, and PostgreSQL. The entire service will be launchable via a single Docker Compose command. I will also include secure API key handling, input validation, error management, a Postman collection, PEP8-compliant code, unit tests, and a Markdown setup guide. A few questions: 1. Do you have a Google AI Studio API Key and a preferred Gemini model available? 2. Is there a specific schema or validation rule required for the JSON output? 3. Should only the request metadata be stored in PostgreSQL, or the model response as well? 4. Will you provide a GitHub/GitLab repository, or should I set up a new one? I will deliver a testable, secure, and easily integrable PoC that fits seamlessly into your team's existing technology stack.
₹12,500 INR in 7 days
0.0
0.0

Hi, here is how I'd build the PoC in Python: - FastAPI: POST /v1/generate (text -> structured JSON), POST /v1/describe-image (image + prompt -> JSON), GET /v1/requests/{id}, GET /health, OpenAPI docs. - Google Gen AI SDK (google-genai) with your AI Studio key from GEMINI_API_KEY in .env (never committed). Structured output through response_schema built from Pydantic models, so responses arrive as validated JSON, not parsed free text. - Images: MIME and size checks; inline bytes for small images, the Files API above the inline request limit. - Reliability: timeouts, retries with exponential backoff on 429/503, clear error JSON, and safety-block reasons returned instead of empty answers. - Postgres via SQLAlchemy + Alembic: request id, endpoint, model, prompt hash, input/output token counts, latency, status, timestamp. - docker-compose (api + postgres with healthcheck, migrations on start): one docker compose up brings it online. - Postman collection, Markdown guide, pytest with the Gemini client mocked (passes without a key) plus one live smoke test, ruff/PEP8 clean. First milestone: the skeleton running in Docker with the text endpoint working end to end and tests green. I haven't delivered a Gemini project for a client yet, so I'll share that milestone before you go further. Which model should it target (Flash for speed/cost, or Pro)? Do you have a JSON schema for the responses, or should I propose one?
₹17,000 INR in 4 days
0.0
0.0

One "docker compose up" that brings the service online and fires a real Gemini request with your key: that is the acceptance test I'll build to. You get a Python REST service (FastAPI) with two demo endpoints, plain text and image-to-text, returning structured JSON, with request metadata stored in Postgres. Also included: a Postman collection, a Markdown setup guide, and a pytest suite that passes PEP8 linting, all in a repo on GitHub or GitLab. Fixed price, 5 days, one round of revisions within 7 days. Which metadata fields should each request store?
₹12,500 INR in 5 days
0.0
0.0

Hi, Your PoC is well defined and can be delivered as a compact Python service rather than an overbuilt application. I propose the following implementation: - FastAPI REST service - Official Google Gemini API integration - Text-prompt endpoint - Image-to-text endpoint - Structured and validated JSON responses - PostgreSQL metadata storage - Environment-based API-key configuration - Docker Compose for one-command startup - Postman collection - Light pytest coverage and PEP8-compliant code - Setup and usage documentation Our relevant experience includes developing production Python/API automation, webhook-driven workflows and persistent operational-data systems. We have also built a live opportunity-intelligence platform connecting external APIs, cloud processing, database storage and a desktop application. REST, database, structured-response and containerised-service architecture required here is directly within our development capability. Please confirm whether PostgreSQL should store only request metadata, model, status and token usage, or also the complete prompts and Gemini responses. let us open a discussion. we are ready to work. Best regards, Afia Nawaz Accura Engineering (Pvt.) Ltd.
₹12,500 INR in 7 days
0.0
0.0

Hi, I can deliver a small, extensible Gemini multimodal PoC as described: a clean REST service that authenticates via Google AI Studio, accepts text and image prompts, returns structured JSON, and stores key metadata in Postgres—all runnable with one docker compose up. What I will ship in 3 days: • Dockerised Python (FastAPI) service + Postgres via docker-compose • Auth wired to the Google AI Studio / Gemini API key you provide • At least two demo endpoints: plain text → JSON, and image+text → JSON • Metadata persistence in Postgres (request id, model, latency fields, timestamps) • Postman collection for the sample calls • Concise Markdown setup & usage guide • Light unit tests; lint-clean (ruff/PEP8) Relevant work on my profile: AI/API integration tools and FastAPI backends. Happy to share a short walkthrough of a similar REST+AI layout after we connect here. I prefer Python so your in-house team can extend it easily. Your acceptance criteria (Docker launch, two live Gemini endpoints, lint/tests pass) are the definition of done. Quick checks: 1) Do you already have a Google AI Studio API key and a preferred Gemini model (e.g. gemini-1.5-flash / pro), or should compose leave a clear env placeholder? 2) Any required JSON schema for structured responses, or is a sensible default shape OK for the PoC? Please fund a milestone on Freelancer before I begin so we both stay protected.
₹16,500 INR in 3 days
0.0
0.0

Hi, I'm Chien. I can build your Gemini proof-of-concept as a small Python REST service that starts with one docker compose up and answers a real Gemini request with your AI Studio key. What you get: 1. A FastAPI service with two endpoints: text prompt to JSON, and image plus prompt to JSON. Each response follows a fixed schema your team can rely on. 2. Postgres in the same compose file. Every request stores its metadata (endpoint, model, prompt length, latency, token counts, status) in a table created by a migration. 3. Your key and the model name live in a .env file. Pasting your key there is the only setup step, and you can switch models without touching code. 4. Unit tests with the Gemini client mocked, so they pass without any key, plus one live smoke test that runs only when a key is present. Ruff/PEP8 and the tests both pass in a single command. 5. A Postman collection with sample calls for both endpoints and a short Markdown guide for setup and use. How I check it before handing over: a fresh clone, docker compose up, lint and tests, then both Postman calls against Gemini with the key you supply, and rows visible in Postgres. My bid is INR 20,000 fixed, delivered in 4 days from award. The code goes to your GitHub or GitLab repo. One question: besides the request metadata above, should the service also store the full prompt and the model's answer, or keep only metadata for privacy? Chien
₹20,000 INR in 4 days
0.0
0.0

Hi, Your Gemini POC needs a small service that authenticates via Google AI Studio, takes text and image prompts, returns structured JSON, stores metadata in Postgres, and comes up with one "docker compose up" — that's the setup I do hands-on. My approach: 1. Python REST service with Google AI Studio auth; two demo endpoints (plain text + image-to-text). 2. Postgres for metadata, Postman collection, light unit tests, PEP8/ESLint clean. 3. Docker Compose containerisation, repo pushed to GitHub/GitLab, concise Markdown setup guide. 4. Verify acceptance: Docker launches with no manual tweaks, both endpoints hit Gemini, lint and tests pass. Timeline: 7 days — days 1-2 skeleton and auth, 3-4 Postgres + endpoints, 5-6 Docker, tests, docs, day 7 handover walkthrough. Price: ₹25,000 fixed, mid-range in your ₹12,500–37,500 budget, released on milestone once acceptance passes. CS background, I do this hands-on via remote session. License: I'll check the tool's license first; some are not MIT, and commercial or SaaS resale may need a commercial license. API costs: model and API usage fees are billed to your own account; I set up your keys and don't resell usage. Copyright: cloning other people's videos, ads or likeness carries copyright and likeness risk, so use only content you own or have rights to.
₹25,000 INR in 7 days
0.0
0.0

Kanpur, India
Member since Sep 14, 2026
₹750-1250 INR / hour
$1500-3000 USD
$30-250 USD
₹750-1250 INR / hour
$25-50 USD / hour
$1500-3000 AUD
₹600-1500 INR
$15-25 USD / hour
$15-25 USD / hour
$15-25 USD / hour
₹12500-37500 INR
₹12500-37500 INR
$30-250 USD
$30-250 CAD
₹600-1000 INR / hour
₹12500-37500 INR
₹12500-37500 INR
₹750-1250 INR / hour
£5-10 GBP / hour
₹750-1250 INR / hour